Bubbles, panic and crashes: what we can learn from ants and bees
Bubbles, panic and crashes: what we can learn from ants and bees
The explanations offered for the recent economic crisis constantly refer to three concepts – trust, networks and contagion – which rarely feature in economic models. If we think about the players in the system – individuals, companies and banks – as subjects who interact constantly and much more “locally” than is thought, then perhaps ant and bee colonies can suggest some interesting models for analyzing modern times.
good afternoon thanks for being here with us thanks for joining us sitting next to me and professor Sherman professor emeritus at the xmax a3 universe and the university pulses and in the rector did they called the Sun social professor Sherman has obviously strong ties with both France and England but also with Italy because he studied in Bologna in Johns Hopkins University in the 1960s he carried out some research work and he worked in the toad actually at the European Institute in Florence and this Anton school in Pisa so he will be able to understand your question in Italian as world without simultaneous translation and he and what is very interesting in his work is this empirical micro economic approach which was which micro economics which was his a starting point if I got it rightly and you will be able to read some of this in his a forthcoming book a book which will be published in London in after some it's titled complex economics individual and collective rationality with an explanation starting from the fish market in Marseille and also research was carried out in an corner in an Italian token signe and is this an empirical approach he followed in the Bank of England he is a research fellow actually he visited the traders and they were very happy because thanks to him they could better understand what they were doing in trading rooms thanks to his analysis and this empirical and dynamic approach I think is very commendable professor Kaman hates static things things that never change and he discussed the processes dynamic evolving things as you will see in his presentation and he also focus on the concept of trust relationship meant a you donkey of a circus and actually he for us journalist after Li we journalists a very ignorant but we write nevertheless and a lot of things he helped us abandoning the let's say the stereotypes of economic journalism and in his book he clearly states that there is no market the market thought the market feared by reading professor Chairman's book you understand the plurality of agent so you don't have a single market and his evolutionary approach after he took inspiration from physics as social sciences biology because he has he will be telling us economics is a complex system and as such you have to study it without simplifying too much Paul the philosopher lon hollandaise he reminded us about that philosopher Ellenville who in the 18th century wrote a fable about bees which was very cause of lots of surprise then and there is this idea I mean there is something seems and very vague similarity and I'm very happy I'm looking forward to listening to him what his opinion is about economic events in the light of the present Christ grandchildren Paul Davis whose army tally needs your perfect okay new Italian motor topper in perfecto para fairy totally present that shown in italiano no no my studio at italiano imparato l'italiano Karenina leg our economy Cheadle Bellagio a dong-gwon Italian Emoto primitive although not process spear having in Italian Tequesta school Romania thank Tito query for inviting me to this wonderful event I'm still surprised by Italy in the sense that we have thousands of people who come to a conference on economics thousands of people go to a festival of philosophy and yet if you look at why you know you can't believe that this is the same same society that has these tastes it's always a pleasure anyway it's always a pleasure to come to Italy and when Oscar Wilde was asked when he came back from the United States did you enjoy yourself Oscar he replied well there was nothing else to enjoy was that and so the it's not true of Italy it's a pleasure to be here so what I want to talk about today and I must say that the title it was somewhat imposed on me by Tito but I would like to start by asking two questions to what extent should be economic crisis and you will see soon why we're talking about ants and bees cause us to rethink our economic theory and secondly do economists bear any responsibility for the crisis in other words some people claim that economists have a big responsibility and what Paul de Grau said was clearly the financial crisis is not only due to the delusions of macro economists the delusions were quite widespread amongst bankers supervisors policymakers yet society expects the community scientists to be less prone to delusions in the rest in that sense the responsibility of the economics profession is crushing somebody else I won't go through all this in the Financial Times said no no no that's not true trying to forecast what happens in the economy is like trying to forecast what happens with the weather extremely difficult we're getting better at it but don't blame the economists I think that's too naive of you but the argument that we might be responsible for what we do is a very long-standing one after the Second World War there was a long debate about what was the responsibility of physicists for the nuclear bombs the people who worked on the nuclear research did they have any responsibility for the use to which those bombs were put and question is do economists have any responsibility for the use to which their models are put of course you can understand that if I ask that question I believe that the answer is yes so my basic claim is that we have been building for too long unsound models which are used for policy and this is not just simple harmless academic research too many people developed and acted according to a worldview which is unjustified and what you hear over time is a crisis was caused by excesses by people behaving badly and so forth but these are in fact parts of the economic system we can't just say well these are just outside things that disturb the system and I think we were not guilty of not actually forecasting when the crisis would happen but we were guilty of building models in which crises cannot happen so if you hear people explaining today's crisis first of all they talk about a crisis of confidence and at the heart of that problem is for example Greece you hear that we no longer have confidence in Greece and in the Greek government therefore we don't have confidence in Europe but confidence is not something which is included in most economic models then we hear explanations that the network of banks broke down but networks of banks don't figure in most macroeconomic models we are told it's a problem of trust between individuals which broke down that's not part of standard economic models there was a problem for contagion and that again is not part of standard economic models all of these features are typical of complex systems but not of standard macroeconomic models and yet we became very confident in our theory after only about 20 years of calm what did Robert Lucas the president of the American Economic Association say the central problem that depression prevention has been solved Ben Bernanke said he wrote papers celebrating what we called the Great Moderation financial markets are no longer volatile 2003 also since then of course people have changed their tune but I think Lucas still believes that we have managed to master crises despite recent evidence now most of us would like to know how can the economy get into a downturn like the current crisis our question is our economists trying to build models to explain that or do they just offer explanations for that and carry on with the same models that they had before and my view is that we start we carry on with the same models but we started with the wrong basis we start from the idea that we look at an individual who's isolated and who only interacts through the market and yet we're taking away the most important feature of the economy which is that people are interacting all the time with each other and we should view therefore that the economy more as a complex adaptive system now most economists will tell you are but an economic model is not good or not sound if it doesn't have sound micro foundations what do we mean by that they said scientific if we have at the basis rational optimizing individuals and of course for many years this has been from Pareto onwards this has been criticized but nevertheless in most macroeconomic models they're based on rational individuals who optimizing and what is worse we characterized the economy as behaving like such an individual in other words the passage from an individual to the market we just think of the market has been one big individual and yet as Mark Twain said this is all called scientific but as he said there's something fascinating about science one gets such wholesale returns of conjecture out of such a trifling investment of fact now twenty years ago a little bit more than twenty years ago I had a correspondence with Bob Solow the Nobel Prize was famous for his work on growth and as he said my view of the way economists actually do behave coincides with yours I wholeheartedly agree with the point that economics self-destructs in part because because we insist on supposing that everywhere and always individuals maximize pure individualistic preferences subject only to technological legal and budget constraints this is a transparently false assumption and the Brotherhood by that he means economists expands vast ingenuity trying to account for facts within this silly framework there are at least two of us however I am afraid 20 years later I'm not sure that there are many more than two of us and I think the modern macro economists have continued to build war a more abstract and mathematically sophisticated models but these models do not contain the possibility of a crisis and they bear no perceptible relation to reality and what Bob Solow has to say now some 21 years later is maybe there is in human nature a deep-seated perverse pleasure in adopting and defending a wholly counterintuitive doctrine that leaves the uninitiated peasant wondering what planet he or she is on and why do we insist so much on rationality why are we so attached to what we call rationality remember that what economists call rationality is not necessarily what the average person means by rationality and I think it's really for mathematical convenience to allow us to build our models and not because we believe that that's how people behave the assumptions we make are not testable and many people have said that and our assumptions do not allow for people's preferences even to change over time an economist believes that people have preferences they know what they are and they don't learn anything so they used to be an advertisement in the UK which said I don't like Guinness that's why I've never tried it and so everybody said that's a stupid investment and but of course for an economist that's perfectly normal if you have your preferences you don't need to try things to know whether you like the model but that seems to me patently absurd and we do not allow in our models for the influence of one person on another the fact that I do something because my neighbour does something because somebody at work tells me something that we don't allow for and we don't allow for the influence of emotions and yet we know a lot of work has been done to show that depending on people's emotional state they changed the way in which they think and reason and even in financial markets where people are supposed to be very reasonable very rational the people's emotions governed to a large extent the way in which they reason in the way in which they act and we have evidence from neuro economics that in certain I give you a very simple example it's what's called the ultimatum game and in this case you put one person has to decide how to divide ten euros which are put on the table so that person says nine for me and one for my partner and now the partner can say I accept if he accepts he gets one and the first person gets nine or he can say I reject then both of them get nothing now an economists will say that the person must then accept if the amount is positive why would he refuse one euro even rather and only get zero but in fact we observe that people refuse partitions which offer the other guy less than four euros and we know that when they refuse what is activated is the emotional part of their brain and not the calculating part of the brain the limbic part comes and you think no why should this guy get six and I get four I'm prepared to take nothing because so this is not the way an economist would think about this so one way out of this is to say well why are we assuming these people are so rational and when argument is well people simply have learned to be rational they don't optimize but they've learned to use rules which are satisfactory and so as time goes on they learn to use the good rules and so they behave as if they were optimizing and lucus the same person I referred to he said we can safely assume that people do behave as if they're optimizing when they drive a car they don't do the calculations but they managed to drive pretty well so they learn but the problem is what happens if the environment you're learning about contains other people who are also learning so in a market you're trying to learn but so are the other people it's it's not clear who is learning you were the environment let me illustrate that point you see this rat here now he's pushing a button marked food and a scientist believes that he is training that rat to push that button marked food and he gives him food but the rat believes that he's training the scientist to put the food in there when he pushes the button so the rat is saying to his friend oh I think he's learning so you see the idea is it's not clear who's doing the learning you know when you have many people and they're all learning about each other so I think that rather than trying to stick to this very basic assumption about how people behave perhaps we should rethink the whole structure and I'm not alone in this I noticed that ben bernanke now says the brief market plunge he refers to an event on May 6th was just an example of how complex and chaotic in a formal sense these systems have become what happened in Stockman is just a little example of how things can cascade our techno technology can interact with market panic now these are phrases which don't come into our standard models this implies that something else at work other than just individuals making rational calculation now what I would argue is that economists have become too obsessed with efficiency their constant insisting on this arrangement is efficient this is not efficient but I think the real problem in economics is coordination and how we coordinate our activities how people managed locally think thinking rather locally to get the whole thing together and that's when we will come back to the ants nest how do these little ants none of whom has any vision of what's going on in the ants nest managed to coordinate their activities to produce something very complex at the so I would argue that we should view the economy as a complex adaptive system and be less obsessed with efficiency and let me tell you a story just to illustrate why economists are so obsessed with efficiency and other people are maybe less there were three people playing golf there was a priest there was a psychiatrist and an economist and they got very upset because the person in front who had a caddy to help him was playing very slowly so they started to shout from screams eh come on play faster we are wasting our time here and they got more and more upset so finally they sent the priest to find out why he did this guy in front even with a caddy to help him place so slowly and the priest came back very very ashamed and he said you know why this poor person placed so slowly it's because he's blind and he takes a lot of time he said I feel so bad every Sunday I explained to people how to live with other people and Here I am shouting at this poor person and he turned to the psychiatrist and said Joe what do you think and joe said well this is terrible I have all these people lying on my couch explaining to me with their problems and I try to help them and here am i shouting at this person this is terrible and they turned to the economists they say Bob what do you think and Bob says I think this situation is totally inefficient this guy should play at night so you see that economists have a very different vision of the world from other people let me show you another thing which is economists design mechanisms which are supposed to achieve efficient results for example economists are very proud of designing auctions which achieve the maximum revenue for the person who's using those auctions and that let me if I am lucky show you ah now I want to mention just a few things about complexity but I will pass over those very quickly because some people say when they hear the word complexity complexity this is just a fad in economics people just interested in complex systems because it's a fancy word but I think it's not and we have this these following features that compose of interacting agents and these systems the people are very simple in complex systems and the fact that people interact means that how the economy behaves how the market behaves is quite different from how the individuals themselves behave or intend to behave and doesn't that make you think of ants and bees so question is I think I've lost one of my nice movies in here but anyway so the question is why are a Gregorian individuals here's the question why is it that the group behaves differently from the individuals within it why do revolutions happen it's not because the average individual was sitting at home solving his complicated optimization problem and then he thought the only solution is to get out and revolt and so since everybody was at home solving the same problem they all get out and revolt together that's not really how revolutions happen revolutions happen because people communicate they interact they influence each other and finally they get out this ISIF Newton said I can calculate the motion of heavenly bodies but not the madness of people you see here is a group of people who are supposed to be rational carefully operating individuals who are doing calculations unemotionally influenced by them around you can see that that's obviously these people could clearly completely calm and completely uninfluenced by what's going on around and I think you can see there's a contrast between our models and how people actually behave now why shouldn't we treat the aggregate like an individual well let me give you an example my B's example and in the B's example what happens is the following I want you to notice the difference between the example of the representative B and the actual reality so B's that's these people how do they keep their temperature the temperature of their hive constant what they do is they beat their wings and they push air through the hive when the temperature gets higher it needs more air so any sensible macro economist will tell you that the average B as the temperature rises beats his wings faster and faster but if you stand beside a higher than listen to it if that were true when the temperature went up first of all you'd hear and then as the temperature got higher and higher and all these guys are beating their wings faster and faster but you don't it's always the same sound it's the same level of sound what happens is that different beasts join in every B has a threshold of temperature at which he joins in but they all beat their wings at the same speed so the temp the sound never changes and so these individuals are extremely simple they're on or off and when they're when the temperatures high there lots of them who are armed and only a few more often when the temperature is low there are very few who are on and lots are off but they have very simple behavior collectively they have very sophisticated behavior as the temperature rises the amount of air driven through the hive changes in a nice smooth way but it's not because the individuals are sophisticated and that's the point so where does the difficulty with a standard economic model come from well as I said the economy is made up of individuals who in directly and systems like that don't have aggregate behavior which corresponds to the average behavior of individuals what do I mean by direct interaction well people do in fact interact with each other they exchange information with each other it's not true all the information we get simply comes from some central source people influence each other they modify their expectations if I meet people who tell me that the price of Apple shares is going to go up I may be influenced by that people imitate each other when they see somebody's who's successful they look and they see someone like Frank Paul who's obviously successful they say ah well I maybe I should behave like him and people trade with each other the most standard models we have no explanation of who trades with whom who sets the prices and so forth so people are interacting directly with each other now there is one branch of economics in which people do interact directly with each other and it's game theory which you may have heard of and in game theory people do very sophisticated things they think about what when they act what other people will do as a reaction so instead of just being simple isolated individuals they now think if I do this what will he do but if he does that what should I do and so forth so the situation becomes much more complicated so each person is thinking about what the others are doing and he thinks that the other guys are clever too so this is very different from what happens in the standard model I guess so most people here are familiar with at least the idea of game theory and John Nash and so forth but what we have to do for this is increase the calculating capacity of agents if we use game theory we believe they're all acting strategically all thinking about what the other people are going to do and that of course is very complicated just think of the following problem you're driving on the outer strata down from Trento to Verona and you see that it's packed with people and you look andon than it he's the road next door you notice that people are driving with no problem no traffic jam so you say to yourself tomorrow I'm gonna take that rabbit and then you think you sit huh wait a minute all the other guys on the artist are they're very smart so tomorrow they're all gonna be on that road so I'd better stay on this road then you say but wait a minute let me think about that again because they're also very smart they can make that step so they'll be on the artists plan so maybe I should go on the other road but they're very clever they didn't they'll make that step to and so you see that this will go on forever now this is what's called in game theory the common knowledge problem and it's what makes game theory rather complicated but it was seen this problem long before it was settled as an academic problem by somebody else Charlie Schultz and Charlie Schultz already analyzed this problem and here you can see those of you who are familiar with American culture know that every year Lucy does this she offers football Charlie says hey look what I have which could I interest you in a little kicking off practice and you can see Charlie's expression I'll hold the bar all Charlie Brown you come running up and kicking he's okay it's a deal I know what she's got on her mind he says this happens every year by the way he says every year she pulls the same trick on me she Chuck's the ball away just as I tried to kick it but he says this time she knows I know she knows that I know she knows I know what she's going to do so I said I'm Way ahead of her and of course what happens and what does she say I figured you knew that I knew you knew I knew that you knew I knew you knew so I had to jerk it away and the point is she took the reasoning 2 n plus 1 steps she went one step further than Charlie Brown of course if they had both done the reasoning properly they would never have stopped but you see that this problem means that using game theory is a very complicated way trying to analyze what people are doing so although game theory is a wonderful weapon it's not a good way of analyzing standard market problems now what would be a better way well think of a world in which agents use simple rules and interact with those around them they learn from and about those with whom they're linked and if we take this view what are often called externalities in economic economic when I do something that has an effect on you they become central to the problem and not just some imperfection in the model when people have an impact on what other people are doing when people by polluting influence other people's lives and so forth all of that should be central in our models and not just added on as an extra consideration and once we accept this then we have to explain how do people interact through what network to the interact who interacts with whom and we have to take account of that network of relations which governs the evolution of the economy and understanding the structure and evolution of this network is crucial to understanding macroeconomic phenomena without understanding the network of relations between banks between firms between governments you cannot really understand what's going on and yet it's absent from our standard macroeconomic models we have to think more like these people and these people as you can see hence and I'll come back again to why they're important but you would never try to predict the behavior of an ant nest as I said before from the behavior of the representative ant so let's think about bubbles crashes and panics and how that's related to these ants and bees now remember that what we usually told when we're looking at bubbles or crashes or a crisis is we look for who was responsible but as Voltaire said in an avalanche no single snowflake sees itself as responsible and I think that's the situation in fact there is no individual or no group of individuals who are directly responsible let's look to take an example at financial market models and financial market models always the same building blocks they're very simple in fact even though they are mathematically sophisticated what are the building blocks agents in these models in these markets they have a way of forecasting future prices once they forecast the future prices of assets then they decide how much they want to buy how much they want to sell once they do that then the new price of that asset will be determined now that price in turn will now change people's forecasts about what's going to happen but if the price has now changed people think about what's going to happen and that's how the models were people make forecasts they act on the forecast that changes the price that changes people's forecasts and so forth now most financial markets are based on the idea of what's called the efficient markets hypothesis and you must have heard of that or many occasions and the idea is very simple in an efficient market all the relevant information is contained in prices you never have need to look anywhere else for information all the information you have already and the basic argument comes from the work of bachelier at the beginning of the 20th century who tried to show that prices in a market must follow what he called a random walk that is they can't be predictable if there was some predictable component in prices of assets then somebody would use that prediction to make money so there cannot be anything predictable left and he made two arguments one is that I should I can't know from current prices what future prices will be because otherwise I could make a profit and secondly he made the argument that the changes in the prices would be like as if they were drawn from a normal distribution you just think of you've got you might have a big change but that would be less likely and there will be a nice normal distribution and so this whole hypothesis was developed in bash liaised thesis but his advisor or replan carry a famous french mathematician he said i'm sorry but unfortunately i don't think you should take this work seriously because unfortunately people do not look at their open information which comes independently and then act on it people look at other people they behave like sheep they must on the penny off people imitate other people in markets people do not behave in this way that they independently get information and then act on it people follow each other this herd-like behavior and of course everybody said poor old Ashley he developed this wonderful tool which was used by many people in financial markets and never lived to see that but he also did not leave live to hear Alan Greenspan in 2008 saying this whole intellectual edifice collapsed in a summer of last year but from the outset as I said when Kerry and others argued that the basic assumption was false we know that from the empirical evidence and yet all the basis of modern financial theory is built on that so how come that the people who made the things have used the derivatives and so forth all based on this underlying hypothesis continue to do that despite the empirical evidence and despite the theoretical arguments black shows the famous option price is based on that argument why did we persist well really we're like the people the man who's searching for a coin under a streetlamp and when somebody says to him did you drop it under the street lamp he says no I dropped it over there but over there is dark I can't see anything so I look for it under the street lamp and so you look for the place where you have a bit of light where you can use your models but you don't look where you actually drop the coin but farmer the great pioneer of efficient markets pointed out remember what this is used for the efficient markets hypothesis it's used for many things but in particular is used to justify the idea that the more you diversify your portfolio the less risk there is and that's a very important assumption in modern finance and it's used to justify the extension of financial markets but farmer himself pointed out that that hypothesis is not justified if you don't have the basic underlying hypothesis of normality we don't need to talk about technicalities but I think you've got what I mean but remember that that hypothesis is the basis of the construction of financial derivatives the warren buffett said in 2003 before the crisis in our view however derivatives are financial weapons of mass destruction carrying dangers that while now latent are potentially lethal that was in 2003 five years four years before the crisis started why do we do it well it's inertia you persist with things you know how to do and in the financial markets people have persisted with pricing options in that way because they know how to do that they can solver in the economics case we continue to build models even if they don't have crises in them because that's what we know how to do but as Menken who is cited by Paul Krugman said there's always an easy solution to every human problem neat plausible and wrong and that I'm afraid is the fate of the efficient markets hypothesis but where does it go wrong why does it go wrong well look at these people these people are looking up at the sky what happened was a twenty students went into Times Square and he stared up at the sky after a little while there were a thousand people staring up into the sky why if the first people are looking up in the sky there must be something to see you know so we better look and see what's up there so after a while you have a thousand people staring up at the sky just because they believe from watching the other people that there must be something to see up there and that's exactly what goes wrong in the efficient markets hypothesis let me explain to you with a simple example how rational individuals they're not irrational wind up with a very inefficient result as a result of this behavior now here I'm going to give you an example think of a restaurant now that two restaurants a and B okay and individuals know something about these restaurants they have a private signal that is their friends went there recently and that's very reliable ninety percent of time that tells you the truth and they have a public signal as the geet Mishler from three years back which is not so reliable 55% of the time it tells you the truth so if you get a signal you should look at your private signal first but let's suppose that the restaurant a is objectively better but the public signal makes a mistake and says B is better 90% of the private signals as you would expect say a is better so you would expect everybody to wind up in restaurant a and if they held up their information held out the signal then that's is what would happen but it's possible everybody winds up in B now how can that happen well think about the one person or the ten people who got the wrong signal they got the wrong private thing it's it going to be one of those guys turns out he looks at his privacy going to be looks at the public signal going to be and he goes into B now along comes the next person who got the right signal private signal go into a oh okay I better go into a because that's very reliable then he looks nice ah but is somebody sitting in restaurant B how can he possibly be in restaurant B it must be because he got the private signal that B is better but his private signal is just as good as my private signal so I better ignore my private signal and look at the public signal Oh public signal says B is better so now there are two people in B and you see the third person who turns up says how come there two guys in B and so he makes the same reasoning and he goes into B but now you see what's happened is after the second guy everybody has essentially thrown away their information how do we get information from other people because we see what they do with it but if people stop using their information then of course throw their own information away it will never get out into the prices or you'll never become public information and that's where the information markets hypothesis breaks down so the collective influence eliminates private information and this is of course of contradiction with the efficient markets hypothesis here is an example of an information cascade the first person down here is saying got a stock here that could really excel really Excel Excel cell cell now everybody's shouting sell sell sell down at the bottom here you see a person who says this is madness I can't take this anymore goodbye it's it's goodbye bye-bye and now they all shall hire you bye-bye so and then of course you see down the right at the bottom in the right hand corner you see I've got a stock here and so forth it's gonna start again but this is the same idea the people infer information from what they're hearing from other people so what are we trying to do with financial markets why are we so concerned about it what we have is the following problem here is the stock index for the dax the German stock index and you can see that if you believed in the random walk hypothesis the efficient markets hypothesis normally speaking you think of a trend and random movements around well that could be a trend with random movements around it but what we have to explain is that suddenly there's a major term now all the standard explanations there must have been some major news at that moment which caused the market shocked the market and turned it down but there was no major news at that turning point the only news was that Boris Yeltsin became or ceased to become a ceased to be the president of Russia and that's surely not the news that would make all the major stock markets in the world turn round so something else is going on and the idea is what we tried to do is to build models where these sort of changes big changes can happen not because of some big shot from the inside but they happen from within let me give you a very simple example and when they have to go to a food source and you give them two possible routes choose you might say well they have two little bridges and same length of a path so what will they choose well they'll split half and half right and what you find out is and that's not true and it's tend to concentrate on one path and they'll all go one way and then after a while suddenly blew it will all go the other way and then we'll go this way so you say to yourself now what interest does that half of us economists well think of people who are trying to forecast future prices they have two different possibilities one group are what are called Chartists and you heard about this before here they're people who try to extrapolate from previous prices and find out what price will be another group will be what we call fundamentalists they believe there's some fundamental values and they believe prices will come back and now what will happen is that unlike what people often say you won't get a nice distribution of some fundamentalist some Chartists what will happen so for a while everybody will become a Chartist and then they will flip back and everybody will become essentially fundamentalist so the market will move between these two when the Chartists are there bubbles can happen but eventually they will be taken over by fundamentals so likely ants they switch from one path to another now with a bit of luck here here at the ants you see go into the food and coming back now what you'll see after a while is that these guys will start to use one track more and more can you see now now if we ran this for a long time what would happen is that you'd see ants suddenly go to the other path as well but they don't concentrate sort of half and half even though both paths the same length and the food is at the end here's a another example here the ends these are real ends here you see see that poor guys a bit lost but you can see and what's happened is finally they got onto this track in this particular case that's interesting because the other track is longer so they found the shorter path but in the end they were sometimes gone on the longer path so the end switched from path to path now some people say oh yes but wait a minute the way the reason ants do that is because they lay a trail of pheromone and in financial markets people are not laying trails people are talking to each other and meeting each other well ants also do that see that guy explained to the other guy where the food was you know I told that other guy so ants also do that sort of recruiting they also talk to each other and explain what's going on I was told to talk about Henson bees by the way I remember so now in economics typically people will say there is a sort of equilibrium price for example of stocks and shares and sometimes we move away from that but we'll come back to that in the sort of model that I just explained you that is not the case what will happen if the prices will always be moving in the long-run there is some sort of structure in these prices we know what the probability of seeing certain prices is but we will never be able to explain exactly where the price will be they're always moving and so that's a different notion from the standard models and here you can see the distribution of stock prices if you believed in Bosch Lee is model there would be very few movements far away from the mean that would be the blue distribution but in fact if you have Chartists these people who try and extrapolate prices then you get the red distribution the red distribution suggests that these rather while the events are much more likely to happen than is suggested by the blue distribution you see movements of two to four here could happen with the red distribution but are almost certain not to happen with the British bigger movements are predicted in that sort of model and that's exactly what we see and here you can see a bubble in the prices on the first thing and the fraction of Chartists on the right and when the number of people who are trying to extrapolate prices becomes very high then a bubble occurs okay so let me just explain very rapidly and I don't have much time left I think okay so here is a very simple situation and this is before the crisis and what you can see in the first little slide there is the derivatives that were being produced originally there were a lot of cash bonds around ordinary bonds gradually derivatives took over the whole MA and you can see that by 2007 we've got a huge amount of derivatives fancy derivatives based on in particular mortgages I think everybody here knows how the market worked what happened was the banks lend to people who bought houses those house loans were then sold to other banks and packaged together with other loans and derivatives are instruments which contain many house loans now what we began to see and you see in the second part of the site is that this is the rate at which people were defaulting on their home loans so we had clear evidence by 19 above loans in 2006 and issued in 2007 that they were defaulting much more rapidly than earlier loans so we knew that something was going wrong but what happened to the prices of these mortgage-related securities they were perfectly flat until suddenly first of all the worst sort of bonds saw the collapse and then the others collapsed shortly afterwards how can you explain that well you can construct a very simple model in which the following happens I have to decide somebody offers me these mortgages together in a derivative I say do I check on it let me have a look at the underlying mortgages and see whether they're good or not or do I just buy it and try and sell it to somebody else now I look around and see what the other people are doing if everybody is just buying and selling on I don't want to spend my time and effort trying to check on the value of the underlying security so what I do is I simply buy and sell on but if it only takes a small change in the probability that people will default for somebody to start checking when he checks the people watching him sit out but he's checking so maybe I should look at my information from the underlying security to check because he may not buy the asset that I'm buying so when this happens you suddenly get a cascade downwards as people all start to check on the underlying securities when they find out that they're bad the price of these things drops and we have a huge collapse in that market that's exactly what happened once again you see what happens is the people are interacting with each other and as long as they are all adopting this rule don't check on the underlying securities everything's fine but the day that people start to check the whole thing can collapse and that's exactly what happened so once again is the interaction between individuals that makes things go wrong let me come to a next point which is I said to you the structure and the way in which markets are related is very important now here is a long quote from Andy Harlan at the Bank of England but what he essentially says is the following he says when you look at the Lehman Brothers failure and the current epidemic of bank failures what he said was this is nothing different from an epidemic of flu like the influenza epidemic sort of hit us in the past it has the same characteristics you have this complex interaction between the different component parts and gradually they influence each other and suddenly the network of banks collapses just as you have an epidemic that flows through a population and just as you have for example the collapse of an electricity system when the network breaks down so here let me just illustrate for you that the structure the financial network is very important what everybody told you was the international financial network is becoming more and more connected so it's getting bigger and bigger so risk is getting less and less because risk is spread out over this whole thing but the reality is that if you look at these charts you will see that this network was becoming more and more fragile let me just explain to you roughly what this net these things mean look at this first chart the nodes or the little points in the graph correspond to countries the size of those corresponds to the amount of assets of other countries that they have if you look at Japan and the US the size of Japan depends on how much US assets it had and size of Europe of the US depends on the amount of assets of the other people and thickness of the lines between the two shows you how much each other held of their assets or US and Japan what you should see here is that this network in 1985 was already quite connected but the countries in terms of their holdings were not very large 1995 much bigger countries are holding more and more foreign assets and the big countries are getting more and more connected 2005 look at what happened we have huge amounts of foreign assets being held by countries like the US Japan and the UK and these countries are very closely linked now the whole network is becoming more and more connected but that does not mean that it was not more and more fragile what was happening is that the network because you have these big countries holding lots of assets and other countries became very vulnerable a small change in us can have a big effect on other countries and so this network became extremely fragile and so there was an epidemic if you like so looking at the structure of these networks enables banks the central banks to try and understand when things are becoming dangerous or fragile there was all the danger signs which I won't read to you but the result is that these systems are very vulnerable and are very vulnerable to the transmission of problems between nobody planned it that way the system developed like that and so what we need all the time is to watch for signs in the system that this sort of thing is happening now people will say but wait a minute shouldn't you just leave the markets to themselves they will self organize and that was what Vernon Smith suggested in part of his talk and so people have used that as a justification for not interfering with mine but you've just seen as an example of half and then international financial markets became very vulnerable without anybody planning it that way so markets do clearly self-organized but we have absolutely no reason to believe that that's a stable process that when they self organize that that will stabilize things and indeed what can happen is that as the actors within the markets gradually change their then we can get a major change in the whole system and it's not because somebody is guilty of doing something big and nasty because gradually the rules change and the system becomes fragile and suddenly collapses so if you have a system like that you have heard people calling out for we need new rules much more strict we need to go back to the old rules but the point is you cannot design the rules for such a situation for financial markets unless you really have a good understanding of precisely how they function and the problem is that they change all the time these markets evolve all the time so the idea that we can write down hard and fast rules now which will govern the whole of these financial markets for the future is simply not acceptable because we have to constantly be modifying the rules controlling these markets watching what's going on we cannot just write down a nice set of rules and say now let the markets get on with it unfortunately the markets will be very are very unhappy with the idea that there's constant monitoring and constant changes of the rule but I think that's the only way you can regulate a complex system so as I say we it's it the idea we can just set up these set of rules is illusory because we don't have a correct model of how the economy functions and how markets function and that model in itself is always changing so the rule should have to change too we know that markets stock exchanges up all the time changing their rules because people learn how to use those rules learn how to get around them so what should I conclude what we should do is make models of the economy which take into account the direct interaction between individuals this is a central and not a peripheral concern in the financial markets prices are constantly moving and do not settle down to a steady state the economy should be viewed as a system made up of individuals following simple rules and to repeat we're not guilty of not having a mirror to forecast the onset the car crash but we are guilty of having built models in which it could not happen and worse I think our models have been used as the basis for recommendations which have led to widespread misery widespread unemployment and I think we have some responsibility for that how long will it take to change well I think we have to keep an open mind and try and change our models and change our way of thinking but as best Brock said you want to keep an open mind but you don't want to open it so far that your brain falls out how long will all this take well what Max Planck said a new scientific truth does not triumph by convincing its opponents and making them see the light but rather because his opponents eventually die and a new generation grows out that is familiar with it and I am afraid that's true you have to wait for the current generation of economists to die and to be replaced by more open-minded ones and let me just say that if you wish to learn more despite this talk there's a book coming out which you might be interested in if you copy that URL you can get this book for nothing I'm not sure mr. Ragnar should be happy but it's a preliminary version is what I can say and what I would like to do just before concluding if you will allow me is to try and show you the one little film that was missing from my talk and I will try to get it back for you because I think you might quite like this film which is about the let me try I hope I can show it to you Oh says he doesn't want to show it so I think unfortunately maybe I can show you it in another way excuse me just a second let me try and find this close that I will just try to find you this little thing in files because I think it's a nice illustration of how economists claim to have developed very efficient very efficient mechanisms for solving problems and you should be able I hope to see this and let me just put it on if I can on full screen I'm not sure that I can manage that let me try can you can you see that okay so let me just tell you the story there is a big fire in a building and people are wanting to get out of this building and of course the firemen arrived but the firemen only have one trampoline to catch people with so the firemen in charge was obviously full of courses in economics decides to auction the right to jump from the buildings and so now you can see it's very bad quality but I think you will understand the idea ah yes I need to put the sound on where do I have to stop second so we need to have to go back and put the sound on your you'll see they think now we hurry I've turned it off in the in system preferences excuse me a second I'm sorry about this but you will see this end here the terrible things that happen in a minute okay I will show you to you again in just a second as we do yeah okay try some questions like was I try and work it out okay why don't you get Garrett see me Leigh Allen request representation a fashion antiquey Thank You Alan for this fascinating presentation that was really very interesting this is food for thought lots of food for thought correctly now I think you'll be able to enjoy if that's the word this particular she laughs at our kappa dolfinh fantastical see don't question family-friendly so a lot of food for thought impulse including this latest last video there are so many questions maybe the question on restaurant a investor will be just at the end so that we know how to get the right restaurant this evening ballots are from arrived here could you please ask brief and very clear questions can you apply your model to the industrial clusters as you know industrial clusters that you know in Italy it's a important issue so in terms of entrepreneurs starting a new business then maybe talking in the bars the evening with other potential colleagues then move the the following morning open a new business in the same sector or even doing the same thing can you apply your model night like ends as you said yep I think this is a very good example and you heard from Bob Putnam also this notion that if you get a network or a cluster of firms individuals you build up a certain capital understanding of what other people are doing that's why we get things like Silicon Valley and we get these collections of firms in the same segment gathering together they create externalities for each other and the communication is between them and contrary to what is often thought where you say what people in competition constantly there's a lot of complementarity a lot of activity which is results from exchange of information and opinion between people in these groups and so you get in these groups a much higher level of awareness of and ticked off technologies for example as a result of this interaction again exactly as you're saying nobody put out this information nobody publishes it centrally but within these clusters it develops and just as in even in sports teams you know you get a cluster of people who play together and they gradually somehow mysteriously they managed to transfer to each other their capacity and they simply play better than other groups and it's a it's a phenomenon which is very important is grouping together people and transfer of information amongst them demand a straight other questions on any point I have a question actually your conclusions I think are right but very pessimistic as well because you say these complex systems are difficult to control and therefore we have to adapt rules continuously change them continuously do you think that regulation as in the g20 hmm I mean having stricter regulation is this useful or is this just smoke in the eyes of this present that is this fight between these two types of thought on the one hand that there is a group of people who say basically we have to set new strict rules maybe and after that markets will be left to themselves and they will work properly visit the idea of some countries of like Germany having strict strict rules without intervening but having very strict rules and the other group like myself believing that we can't really have such a rigid stiff structure and then say stop with these rules and we need police sorry with these rules is they are enough and we just need police to enforce them no we need more and let's say flexible rules so this is what these other people believe we have need flexible rules and as soon as we identify weakness science we change these rules it's like reserve requirements as in reserve requirement so it's like a doctor always checking the patient's health and modifying treatment accordingly obviously you can't give a medicine for five years you modify your your drugs your medicines and so there are countries that want rules without intervention on the markets and other countries who want a more flexible system other questions Awards were all sorts of modern policy analysis these DSG models which are widely used in the ECB for example even on what do you think of this modern tools of economic analysis do you think are completely - to be put in the rubbish bin because I see this standard tools of economic analysis you mean use like dynamic general equilibrium model do you think because if I read in you are correctly in your slides is also very strong critics of this DSGE models yeah I think it's it's not a polite thing to say I mean I think let's say if we were talking just together you and me I would say well we should probably put them in a wastebasket but since this is a public audience I would say we should pay much less attention to them I think that the structure of those models this structure those models is fundamentally wrong it's a static structure even though they're called dynamic they're not really dummy structure those mothers is is static they are essentially equilibrium models where you're in an equilibrium and every now and then there's a shock which knocks you off but you come back on the path and I think that's not what the structure of the economy looks like so to be honest I think that we have to rethink the structure of these models and possibly use some newer models from physics as somebody said recently our standard model in economics is based on 19th century physics classical mechanics so since we spent the 20th century using 19th century physics perhaps in the 21st century we could use some 20th century physics and I think we should learn lessons from dynamic analysis in statistical physics for example we can learn lessons from that and build models which the evolving dynamic nature of the economy so I think the answer is yes that's where they should go this we understand you it's very difficult to make models and one has to be very careful with these models and follow certain rules with you our time but my question is did you come to the conclusion that certain areas of financial transactions where one should economies should abstain from stay away from completely because the risks that the system runs out of control is too big or are you saying we will that is going on at the moment only we will have to be more careful and follow other rules in forming models now I think I I would not say that we should simply let things go on I think you're quite right we can identify clearly now after the crisis some parts of the financial system which seems simply to lead to general fragility for example very high leverage in financial markets makes people extremely vulnerable to a downturn allowing for the development of a non-bank sector where banks place their assets and therefore don't seem to have assets which are vulnerable that also is very dangerous and I think either we have to do something about diminishing that sector or including it in the banking sector and applying the same rules to that sector as it is we have two types of banks at the moment and banks easily transferred their assets to places where they were no longer controlled that sort of thing I think has to be restricted so I think my idea is that as we watch the evolution of the economy it shows us certain things are harmful certain things are likely to cause a major downturn in the economy and those things we have to take care of and so I think we're constantly gonna have to find rules to stop that just a simple example recently we have a sudden downturn of 10% in the Dow Jones why because there was all this computerized trading so we get to some level of prices and at that point the computers have been instructed sell so suddenly lots of people are selling and this is nothing to do with what people are thinking about the market this is simply something that people have put in place so one has to do something to stop that sort of automatic collapse and indeed we've put in place now circuit breakers ways of stopping that trend but we constantly have to look at these from and I don't think that if somebody had sat down before and said let's design a set of rules that that rule would have been in there so the fact is we have to constantly watch and change the rules to see when these things happen at a downturn of 10% in the major stock stock exchange index in the world is enormous so and that was induced by a phenomenon which we had not controlled before so I think we have to constantly be limiting these things so I'm not saying hands-off I'm saying we have to constantly be looking alike the doctor looking at the patient he's got a fever we have to do something about this weight of the fever come from so that ie you understand my microphone I hope something and we need coordination because if we don't have coordination between countries of course people will just run to the country where they don't have those rules but I think we do need to coordinate on certain agreements in the major markets are there are we capable what is very strange is that two years ago we were all worried about saving the financial markets essentially stopping them collapsing and governments poured money into the markets in order to save them and now we're all worried about the financial markets we say well we have to take strict measures because the financial markets won't like it otherwise so somehow all the power has shifted back quickly to the financial markets but I think we shouldn't allow that to happen too much that the financial market should not govern what countries have to do and we have to do take measures to stop that and that's difficult I really do we Kimmy demand demand the Queen two more questions hello an economist but a micro economist somehow well that is what I have seen in fact recently many people doing research on experiments bounded rationality and so on so it seems to me that the major I mean I would say that your pessimistic view relates in my experience much more to the macro so to the study of the country's economy rather than the choices of individuals why is there this gap still I mean if it's true what my perception why is there this gap still between factor people do experiments and you know they try they know that standard micro is not the new classical models not perfect the question is a very good one and what happened was in the 1970s micro economists and general equilibrium people understood the weaknesses of the model and in addition with Carmen Tversky and other people the notions of bounded rationality and Simon came in and now we've developed behavioural economics but macroeconomics somehow decided that the classical standard microeconomics should be the basis for macro and it hasn't taken into account the evolution of what happened in micro and that's a very very strange phenomenon macro has become very rigid very bound to a model which most micro economists realize has many weaknesses so your question is a very good one but why do we get stuck like that I have no idea why is it that macro economists are much more pure my old micro theorists and modern micro theorists they if you look up in the modern macroeconomics textbook we always have an individual a representative individual I mean analyze his optimizing behavior and that modern micro economists know all the weaknesses there have been experiments done we've seen we know about bounded rationality when we know a lot about people's reactions there is behavioral finance but that doesn't get incorporated into modern macro models for example so why that's true I think it's just a question of intellectual capital macro economists invested a lot in those models and they're not prepared to give that up without a fight not a mad Amanda there's been a lot of talk about moving from an Information Age to the conceptual age nowadays and let's say moving away from a left brain approach of based on data and models and information and into a more right brain or whole brain let's say approach and how is that reflected in can you know moving front well Daniel Pink's new new book moving from a left brain to a whole brain approach and integrating more let's say our right brain capacities in to and letting go of the information data model kind of way of making decisions than just in structuring our world and how is that being reflected in economics so if I understand the correction at the question you're asking in a more modern psychological approach would be to consider the functioning of the whole brain and reactionary I think that people there's evidence from neuro economics a cell that people are starting to think carefully about how people take decisions and people take decisions in a much more complicated and integrated way than simply taking data and processing it is that the the yeah and once you incorporate that into financial models that makes you treat agents as behaving very differently than they do if they're just simply individuals who receive prices process them and then make their decisions people have very complicated reasoning and Andy Lowe at MIT has done a lot of work on this on trying to analyze the reactions of traders even looking at their emotions as measuring their skin their perspiration their heartbeats and their brain activity we're far from really understanding how people make decisions but I think we can't ignore the fact that people we should use psychology and the fact that people are influenced also by psychological considerations and that the whole brain is operating in a very complicated way but we don't have models which successfully incorporate that yet one day maybe but maybe we'll never understand the black box you know we do know which parts of it light up from time to time you that simile Alan crystal tempo in she regrets it thank you thank you to you all i really suggest you if you go to have a look at his book and download his book complex economics it's really a clear book in terms of language presentation and you'll learn a lot I can assure you and its food for thought for weeks and months ahead thank you