Benoit Mandelbrot On Efficient Markets- FT.Com 9.30.09

welcome back to the future of investing I'm here in the Metropolitan Club in Manhattan at the global creative leadership summit where I'm going to be talking to benoît mandelbrot who is in many people's opinion a genius he's a mathematician now well into his 80s who was born in Poland spent the war in France and then came here to the US he's best known as the inventor of fractal geometry but he is now perhaps most relevant to us because he was also the first and most prescient critic of the theory of efficient markets which led to the very complicated investment products that crashed disastrously in the last two years causing the crisis he was much criticised for his ideas about markets when he first produced them more than 40 years ago now but in the last few years many people are beginning to think that he was right all along that means that his views on the future of investing are particularly interesting let's talk to him that professor Mann about thank you very much for being with us here today thank you very much for a bit let's start if we could buy if you could explain the where the efficient markets hypothesis comes from where where it intellectually started and what it what it presupposes well you started precisely in the year 1900 when a man named Bashar Yi a French mathematician presented his PhD dissertation called the theory speculation it's a remarkable piece of work which has had an enormous influence on here probability enormous influence on what's called Brownian motion but it was meant to be a theory of the way in which prices vary on markets of securities or commodities etc as such it had the most bizarre history for six years hardly anybody pay attention six years later when the computer became available when people could read compare that theory will with the data the bascially theory suddenly became known people became aware of it and in the gun to rule the roost front of those were trying to model the variation of prices on various exchanges and what exactly does the bash in the a theory they suppose or positive out the way the markets were well it polished markets work in extremely simple fashion as much of a matter of fact it posits that markets are simpler than most physical phenomena which is quite extraordinary strong assumption the it posits that in a certain sense price goes up or down by toss a coin if it has the price goes up a little bit if the stage goes down a little bit and just goes on like that endlessly if you why when the price is high well you will if you use if you buy own prices alone you you you gain but it's a model of a see simplicity which has in in certain sense no parameters if you scale the time appropriately and the value appropriately it's the same theory for everything so it's a it's a complete random walk and the chance of a big move on the stock market is the same as the chance of the same coin coming up heads twenty times in a row for example it that yes kind of idea and a chain saw a sharp discontinuity is zero because the process is assumed to be continuous price varies continuously by by amounts which of course in reality one doesn't have the continuous change because one has finite number of readings but the theory says price varies continuously in produced time right and remembering the day early in 2007 when the Dow Jones Industrial Average fell 200 points in one minute that already begins to sound as though it's not a very accurate portrayal of the way markets actually work but had he actually done empirical work on markets or was this a theory based on assumptions well Bassel ye did not mention any experiments rumor is that he had done some experiments but they are not reported has written about it in other other papers of his but one knows very little about him he is a kind of mystery man in history mathematics he was presented PhD got the doctorate and disappeared for quite a while until much later became professor of mathematics in the provinces one does not know his motivation very well now why do you think he appealed to this generation of financial economists who came along in the 1950s and 1960s and came up with all the ideas that I don't the girded derivatives and the complicated product that that we all got to know about last year when I was present at a time I was close to that yield the reason was that the computer came many things would be actually done tested data became available and one was in a certain sense for the thirst for existing theories that might help and here miracle of miracles in 1900 a completely complete theory was discovered rediscovered in libraries and one could have already a history one could pick out of bascially all kinds of formulas which which answered questions which people are asking themselves now as I understand it the bachelier theory implies that returns follow what we all commonly know as the belly f that most things will be clustered in the middle and they'll be just a few what wider movements at either end of the scale what did your research tell you once you started taking a look at this which is what in the early 1960s that detail is in question which are so completely impossible to draw because they are so thin just a few events it is an extremely long and that record one looks at and seem to be just small steps up and down and then one gigantic step and the gigantic step itself is very often in a certain sense segregated market stop straining straining stops and it assumes a different value the question arises do you take this big enormous jump as part of the process or do you say it's after God depending upon different rules he says after all it fits in battery if you wish because you have put all the big events under the carpet but those big events are actually what matters most about markets aren't they absolutely so what was your alternative idea that you were putting forward four decades ago and why do you think it was not taken more seriously in when it came to drawing up the way in which we invest well more than four it decays across it made quite a sensation it was City sized before to published which is a compliment which one doesn't want to have but which is any case what happened my theory was based upon first of all an observation which I made accidentally I was coming to the lecture at Harvard as a matter of fact and on the board of my host I saw a certain drawing which I was going to use in my lecture so I asked my host but how come you have in your blackboard the drawing I'm going to use he says I had no idea what you're going to say but this this simply refers to price change and the first the distribution of one price change and next one the Joint Distribution it turns out for technical reasons which I cannot explain here that there are two different shapes of this distribution one which is goes this way one goes this way and of course you can't go in between everybody was assuming it was that way right what I saw in blackboard was this way in other words the data were presented to me ideal fashion but your chance and I asked my my my my this chairman who became my friend to give me his card look I went home to my to where I was working at IBM and analyze this data and found that this distribution as I had inferred from for my own reasons represented a situation in which the large price change even though few were so large as to be the most important thing there is so that's utterly contrary to bachelier yes events and in the my my work was very soon city-sized quite sharply of course and but the main critic who wrote a very very nice nicely written story say that well we have known that were these big values but since we didn't know what to do with them we just swept them under the carpet so if you're but if you're trying to model risk management and that's the main use the deficient market theory has been put to his managing risk they're admitting from the start that they're not trying to take account of the biggest risks when managing risk well that is that person did stated very clearly in a in the kekeke of my work in a book that he edited in 1963 or 64 and why do you think that the theory became so popular is it because it made investing look easier than it was that it made the world look simpler than it was or what lies behind it certainly the word simple any word that was the reference specially in terms of influence was not there I mean the work we have occurred anyhow but the same that field bascially had become very important as the Brownian motion and there are books and books and books about it many people knew about it and therefore there was a certain capital of knowledge capital of of wisdom which was available and one could draw upon and on one hand on what other hand my work had one two references were obscure not very close to what we're looking at in a certain sense my work promised the well great amount of work and trouble and effort and the other promised a capital on which one would live for a while so in some sense this is down to academic politics that if if efficient markets were true then we can have build a lot of theories from it and if your theory is true then we have to work a very long time and we may not find anything that we can use at the end of the work well it's exhibit version of what I think happened yes

Channel: valueinvestingpro

Views: 37,015

Rating: 4.9577465 out of 5

Keywords: 2278DBA4, Mandelbrot1, 1254282368543

Id: vxbxXBrOPS8

Length: 10min 52sec (652 seconds)

Published: Sun Oct 04 2009