Thursday, July 14, 2011

Nassim Taleb to Advise IMF on How to Predict Unpredictable Events

Black Swan events, which are  thought of as a metaphor that encapsulates the concept that the event is a surprise (to the observer) and has a major impact, apparently can be spotted after all, if you are doing so for the high-profile IMF.

Nassim Nicholas Taleb, who developed Black Swan theory, once wrote at NYT:
Black Swans being unpredictable, we need to adjust to their existence (rather than naively try to predict them).
But, today we learn he can predict them afterall (for the IMF). Bloomberg reports:
Nassim Taleb, author of “The Black Swan,” will work with the International Monetary Fund on drawing up measures to detect hidden risks in the financial system for G-20 ministers and central bank governors.

Taleb will team up with the IMF’s Monetary and Capital Markets department on a paper for the G-20 to develop ways to apply his method for identifying tail risks, or the chances of low probability, high-impact events, according to a letter from Elie Canetti, an IMF adviser...His “research is part of the fund’s broader efforts to strengthen its early warning capacities, and will be among the many inputs to the IMF’s response to the G-20’s request,” IMF spokesman William Murray said in an e-mail today.
He won't be getting paid. Another seduced simply by being allowed to circle near evil power.


  1. I don't get this. Why in the hell would the IMF ever listen to anything he had to say? They have a long history of doing things Taleb has spoken out against. Is this just window dressing?

  2. Looks to me like Bloomberg erred in using the word "detect" in the first paragraph. The second paragraph makes clearer that Taleb is going to help with estimating the probability of unidentified, high-impact events, which is quite different from "identifying" them, presumably saying what they are.

    Look at it this way: one can make a list of identified events that might happen, assign a probability and calculate the financial impact of each (this is standard Decision Analysis). Then one can lump all the unidentified events into one or more categories with impacts of various sizes, then assign probabilities to those categories. Imprecisely, to be sure, but clearly better than ignoring them. This is Taleb's claimed expertise, assigning probabilities to large, "surprise" (unidentified in advance) events.

  3. It's quite simple plot Federal Reserve Activities add in coefficient for Congressional actions with a multiplier if it's to "help" middle class. When this plot is spiking you know in the future something that an "unintended" consequence is going to take place.

  4. I didn't realize Taleb could even lump black swan events into categories with impacts of various sizes.

    I thought his whole fame came from the fact that he identified that these black swan events exist and we can't do anything about them.

    I enjoyed reading Fooled By Randomness, though.

  5. It dosn't do you any good to estimate the probability of being hit by a train if you can do nothing to actually prevent being hit by a train.

  6. Now go read The Daily Bell regarding ECB's Ewald Nowotny's comments about private sector watchdogs.
    Let all get vetted baby.

  7. E.G. Palmer-- True, if your car is stuck on the tracks and you're handcuffed to the steering wheel, assigning probabilities is an academic exercise.

    But if you are deciding what route to take, or whether to drive at all, accounting for unlikely possibilities may provide insight into your decision.

    Still, getting hit by the train would not be a "Black Swan" event, since that outcome was identified in advance. Getting hit by a falling airliner, or a suicide jumper, or a meteorite would be, if they were not anticipated as possible outcomes.

    As an aside, what constitutes a "Black Swan" event clearly depends on your understanding of a situation. I'm thinking of panics and depressions that are inevitable to the Austrian, and not even on the radar of the Keynesian.

  8. The "Black Swan" was kind of shallow... an entire book based on a thesis that the assumption about independency of random inputs is often inappropriate - which should be evident to any statistician whose head in not completely up his arse.

  9. Scott O- I understand what you mean, I probably should have made a longer comment on that.

    What I meant was that the IMF is choosing to walk on the tracks here, and looking to avoid the results of their own decisions.

    I see any science, including economics as seeking to understand how and why things are as they are. It aggravates me when people start from a position of "this is how we want things to be" and make decisions and pursue activities that clearly conflict with the reality of the situation.
    The IMF wants to use Taleb's ideas to prepare for and ward off events that it's own policies may bring about, rather than rethinking those policies.
    It's missing the point, even as it's missing the idea that a "black swan" event is by definition unforseen, and not something that can be prepared for. If it was, it wouldn't be a black swan.
    The whole thing is nonsensical. It's like a carpenter using a microscope to drive screws because he's afraid it might rain.

  10. In his books he writes about doing risk consulting for places like casinos and the Department of Defense. Maybe his latest venture is, for him, an opportunity for more book material; criticism of the IMF would be heavier with inside experience.

  11. Here;s a good interview with Taleb a couple of years ago that might shed some light. Episode # 85,