7 ms·

Nassim Taleb and Benoit Mandelbrot on the Economy

35 points by nir 18y ago

mkn 18y ago
I've started typing about 5 comments and deleted them all. They don't seem to be saying enough to even warrant comment. When asked about what could happen, there was some talk about a default cascade leading to depressed prices [waves arms] and a grocery store can't make payroll because they can't find a bank to give them a loan secured against their inventory. Now, I don't think he's wrong that we could see grocery stores not making payroll, nor that a default cascade could lead to that. I just think that the arm waving part is what saves us all. If things really do get that bad, then new systems will be formed of necessity. For example, some of those grocery store employees could take a portion of their pay in store credit, micro-lenders could come to the aid of the store (micro-lenders that include the employees), barter will happen, illiquid assets will be traded, and so on. Not that it will be pretty, or that their won't be real pain, but there are ways to eke by. Also, I think the most telling comment of the interview was when Mandelbrot said that we couldn't calculate the odds of a catastrophe like this. Taken to it's logical conclusion, this kind of means that we can't tell if we should even be paying attention to these two on this matter. (I know. Mandelbrot. But even Homer sometimes nods.)
- jamiequint 18y ago
  "Taken to it's logical conclusion, this kind of means that we can't tell if we should even be paying attention to these two on this matter." In that case, it also means you don't know if you should be paying attention to anyone.
- senihele 18y ago
  I'm not sure about your last point. It seems there are some extremely important implications from the determination that "black swan" events are often of incalculable risk. I already posted the link farther down the thread, so I won't repost it, but I highly recommend reading the article - it is written by Taleb and reframes this concept in terms of the current crisis. It really is one of the most interesting high level overviews of the problem I have come across.
- marvin 18y ago
  Mandelbrot practically said that you shouldn't listen to him; not look to him for answers. But in fact, listening to any opinion on the matter is looking for answers. His point is that nothing will give a correct indication about what is about to happen in our current situation. The classical "markets always correct themselves" doesn't necessarily hold.
- colortone 18y ago
  The point that they're making is that we should all expand our perspective on what constitutes possible (or even "reasonably likely") outcomes from this. Maybe I'm misinterpreting you, but I'm not sure if your statement about finding ways to "eke by" includes eating beans in your apartment for 5 years or trying to survive getting drafted and serving in the China-Taiwan war. Not just "getting store credit". Anyways, that's how I'm reading it. Definitely thought provoking. Taleb's piece on Edge.org last month is insanely detailed for anyone looking for more: http://www.edge.org/3rd_culture/taleb08/taleb08_index.html http://www.edge.org/3rd_culture/taleb08/taleb08_index.html
dood 18y ago
Always going to be worth watching, but kind of a shame that they had to talk on such a basic level. I'm sure a 30 minute conversation with those two could be very interesting. Wasn't a whole lot more than them explaining that the system is very complex and hard to predict, and worringly prone to catastrophic events.
- bd 18y ago
  Well, they did mention one fundamental problem which also hints at the solution: the system is too optimized thus it is very fragile, susceptible to amplify even tiny perturbations. Basically, we need to give up some performance and uncouple interacting elements to get more resilient system.
  - dood 18y ago
    Good point, would have liked them to talk about possible ways ameliorate this factor.
- trevelyan 18y ago
  Why are they worrying about the economy when the Arctic is melting?
gills 18y ago
Video here: http://www.pbs.org/newshour/video/module.html?mod=0&pkg=21102008&seg=5 http://www.pbs.org/newshour/video/module.html?mod=0&pkg=...
mdemare 18y ago
I'm dying to know Taleb's thoughts on the crisis, but this video is aimed at 10 year olds. Don't bother.
- MaysonL 18y ago
  There's a good video he gave earlier this year at http://fora.tv/2008/02/04/Future_Has_Always_Been_Crazier_Than_We_Thought http://fora.tv/2008/02/04/Future_Has_Always_Been_Crazier_Tha... Aimed at a somewhat more advanced audience.:-)
  - noor420 18y ago
    thanks for sharing.
  - antiismist 18y ago
    I saw that talk back then ... it is a lot easier to follow his writing than his talking.
- senihele 18y ago
  Really fascinating and insightful essay by Taleb that he wrote about two months ago about the crisis: http://www.edge.org/3rd_culture/taleb08/taleb08_index.html http://www.edge.org/3rd_culture/taleb08/taleb08_index.html Definitely check it out.
kul 18y ago
funny how they labelled Taleb an economist, that must have riled him
- Ben65 18y ago
  I was about to add nearly the same comment.
ced 18y ago
The behavior of economic phenomena is far more complicated than the behavior of liquids or gases. Is predicting the result of a coin toss complicated? Analogy aside, I don't think one can compare the two. Economics deals with a huge (huge!) quantity of information in doing a prediction. If there was a human that could somehow process all the news, weather forecasts, CEO interviews and financial information, I don't think it would be that hard for him to make predictions, even on a global scale. The issue is that we attempt to capture all this complexity through a handful of insufficient abstractions. It works sometimes. Sometimes it doesn't, much like predicting the weather from a barometer. If all that information would be quantifiable numerically (syndicate pressure = 0.34) I'm confident that we would have good models and predictions. Models in economy share a lot of the pain of models in biology. In contrast, for predicting the detail of fluid turbulence or mathematical chaos, the problems are very easy to state, and there is very little information to process. Their mathematical solution and their numerical stability is what makes them complicated.

35 points by nir 18y ago

mkn 18y ago

I've started typing about 5 comments and deleted them all. They don't seem to be saying enough to even warrant comment. When asked about what could happen, there was some talk about a default cascade leading to depressed prices [waves arms] and a grocery store can't make payroll because they can't find a bank to give them a loan secured against their inventory. Now, I don't think he's wrong that we could see grocery stores not making payroll, nor that a default cascade could lead to that. I just think that the arm waving part is what saves us all. If things really do get that bad, then new systems will be formed of necessity. For example, some of those grocery store employees could take a portion of their pay in store credit, micro-lenders could come to the aid of the store (micro-lenders that include the employees), barter will happen, illiquid assets will be traded, and so on. Not that it will be pretty, or that their won't be real pain, but there are ways to eke by. Also, I think the most telling comment of the interview was when Mandelbrot said that we couldn't calculate the odds of a catastrophe like this. Taken to it's logical conclusion, this kind of means that we can't tell if we should even be paying attention to these two on this matter. (I know. Mandelbrot. But even Homer sometimes nods.)

jamiequint 18y ago
"Taken to it's logical conclusion, this kind of means that we can't tell if we should even be paying attention to these two on this matter." In that case, it also means you don't know if you should be paying attention to anyone.
senihele 18y ago
I'm not sure about your last point. It seems there are some extremely important implications from the determination that "black swan" events are often of incalculable risk. I already posted the link farther down the thread, so I won't repost it, but I highly recommend reading the article - it is written by Taleb and reframes this concept in terms of the current crisis. It really is one of the most interesting high level overviews of the problem I have come across.
marvin 18y ago
Mandelbrot practically said that you shouldn't listen to him; not look to him for answers. But in fact, listening to any opinion on the matter is looking for answers. His point is that nothing will give a correct indication about what is about to happen in our current situation. The classical "markets always correct themselves" doesn't necessarily hold.
colortone 18y ago
The point that they're making is that we should all expand our perspective on what constitutes possible (or even "reasonably likely") outcomes from this. Maybe I'm misinterpreting you, but I'm not sure if your statement about finding ways to "eke by" includes eating beans in your apartment for 5 years or trying to survive getting drafted and serving in the China-Taiwan war. Not just "getting store credit". Anyways, that's how I'm reading it. Definitely thought provoking. Taleb's piece on Edge.org last month is insanely detailed for anyone looking for more: http://www.edge.org/3rd_culture/taleb08/taleb08_index.html http://www.edge.org/3rd_culture/taleb08/taleb08_index.html

dood 18y ago

Always going to be worth watching, but kind of a shame that they had to talk on such a basic level. I'm sure a 30 minute conversation with those two could be very interesting. Wasn't a whole lot more than them explaining that the system is very complex and hard to predict, and worringly prone to catastrophic events.

bd 18y ago
Well, they did mention one fundamental problem which also hints at the solution: the system is too optimized thus it is very fragile, susceptible to amplify even tiny perturbations. Basically, we need to give up some performance and uncouple interacting elements to get more resilient system.
- dood 18y ago
  Good point, would have liked them to talk about possible ways ameliorate this factor.
trevelyan 18y ago
Why are they worrying about the economy when the Arctic is melting?

gills 18y ago

Video here: http://www.pbs.org/newshour/video/module.html?mod=0&pkg=21102008&seg=5 http://www.pbs.org/newshour/video/module.html?mod=0&pkg=...

mdemare 18y ago

I'm dying to know Taleb's thoughts on the crisis, but this video is aimed at 10 year olds. Don't bother.

MaysonL 18y ago
There's a good video he gave earlier this year at http://fora.tv/2008/02/04/Future_Has_Always_Been_Crazier_Than_We_Thought http://fora.tv/2008/02/04/Future_Has_Always_Been_Crazier_Tha... Aimed at a somewhat more advanced audience.:-)
- noor420 18y ago
  thanks for sharing.
- antiismist 18y ago
  I saw that talk back then ... it is a lot easier to follow his writing than his talking.
senihele 18y ago
Really fascinating and insightful essay by Taleb that he wrote about two months ago about the crisis: http://www.edge.org/3rd_culture/taleb08/taleb08_index.html http://www.edge.org/3rd_culture/taleb08/taleb08_index.html Definitely check it out.

kul 18y ago

funny how they labelled Taleb an economist, that must have riled him

Ben65 18y ago
I was about to add nearly the same comment.

ced 18y ago

The behavior of economic phenomena is far more complicated than the behavior of liquids or gases. Is predicting the result of a coin toss complicated? Analogy aside, I don't think one can compare the two. Economics deals with a huge (huge!) quantity of information in doing a prediction. If there was a human that could somehow process all the news, weather forecasts, CEO interviews and financial information, I don't think it would be that hard for him to make predictions, even on a global scale. The issue is that we attempt to capture all this complexity through a handful of insufficient abstractions. It works sometimes. Sometimes it doesn't, much like predicting the weather from a barometer. If all that information would be quantifiable numerically (syndicate pressure = 0.34) I'm confident that we would have good models and predictions. Models in economy share a lot of the pain of models in biology. In contrast, for predicting the detail of fluid turbulence or mathematical chaos, the problems are very easy to state, and there is very little information to process. Their mathematical solution and their numerical stability is what makes them complicated.