Ten principles for a Black Swan-proof world

April 9, 2009 12:36 AM

Ten principles for a Black Swan-proof world By Nassim Nicholas Taleb in FT.
posted by jouke (49 comments total)

He is an iconoclast.
posted by Henry C. Mabuse at 1:07 AM on April 9, 2009

Argh. Registration-required link.
posted by koeselitz at 1:16 AM on April 9, 2009

(Link to non-registered version)
posted by Rhaomi at 1:17 AM on April 9, 2009 [3 favorites]

t ^{h a n} k _s !
posted by koeselitz at 1:18 AM on April 9, 2009 [4 favorites]

"Instead, find the smart people whose hands are clean."

Aah, the greatest challenge in about every walk of life. Sure, solve that and you're done.

Governments should never need to “restore confidence”

No, it is despots who never need to restore confidence. The whole concept of democracy is based on hearsay and herd behaviour, hence confidence is a strong part of it.

Is this guy 17 or what?
posted by Laotic at 1:23 AM on April 9, 2009 [3 favorites]

Ah point six is interesting -- "Complex derivatives need to be banned because nobody understands them and few are rational enough to know it."

I wonder what he considers a "complex derivative"? Surely not a Credit Default Swap, the valuation of which is fairly straightforward. It was the improper usage of the CDS, speculation that caused all the trouble, but lots of us working in the field were concerned about the excess.

Without a doubt CDO² or higher moments shouldn't be traded, but guess what? Actual issuance of these structures actually is fairly low - in spite of noise in the mainstream media - because of the complexity. Regulators don't like them, Risk Managers don't like them and even the traders don't consistently like them (I guess selling would be ok, but if you're buying one valuation is tough and if nobody wants to buy them selling sorta dies off).

I don't think he's talking about stock options. We've got housewives trading these things today, even though they've been declared illegal several times throughout history [ .ppt] . The same with futures contracts. We've seen them outlawed many, many times throughout history.

In fact almost everytime we see a financial bubble pop various derivatives are outlawed, only to creep back into usage and go mainstream years later.

For example, stock options were outlawed because we didn't have the math to properly value them. Without a model telling you fair value you've got no idea if you're paying a fair price and then all sorts of speculative excesses arise. Of course Black-Scholes changed all that, even though I'm amazed how many folks I chat with who trade stock options without going so far as to calculate their own prices.

So I wonder what he's considering a "complex derivative"?

Thanks for posting! Especially so the no registration link.
posted by Mutant at 1:46 AM on April 9, 2009 [5 favorites]

Goddamned black swans. Always ruining the economy. Why the hell do they have to do that, anyway? Don't they know that swans don't need money?
posted by koeselitz at 1:46 AM on April 9, 2009 [5 favorites]

Laotic, that's too far out of context for me.
posted by krilli at 2:06 AM on April 9, 2009

Bah, this is better.
posted by GavinR at 2:49 AM on April 9, 2009

2. No socialisation of losses and privatisation of gains. Whatever may need to be bailed out should be nationalised; whatever does not need a bail-out should be free, small and risk-bearing. We have managed to combine the worst of capitalism and socialism. In France in the 1980s, the socialists took over the banks. In the US in the 2000s, the banks took over the government. This is surreal.
...
4. Do not let someone making an "incentive" bonus manage a nuclear plant -- or your financial risks.. Odds are he would cut every corner on safety to show "profits" while claiming to be "conservative". Bonuses do not accommodate the hidden risks of blow-ups. It is the asymmetry of the bonus system that got us here. No incentives without disincentives: capitalism is about rewards and punishments, not just rewards.

Hallefuckinglujah.

I've only read Fooled by Randomness, not "Black Swan", but it's well worth a read. Taleb may be a bit strident, but he's not a nutter or bandwagon jumper or a perpetual doomsayer. He's explained booms and busts and the risks of the financial system long before the current crisis.
posted by TheophileEscargot at 3:08 AM on April 9, 2009 [7 favorites]

Ezra Klien brought up a point worth discussing:

Predictably, [Taleb]'s a bit more focused on avoiding such occurrences than society might want to be. So though his list is a good agenda for becoming "Black Swan-proof," we also need a conversation on exactly how Black Swan-proof we want to be.

posted by jmhodges at 3:08 AM on April 9, 2009

marginalising the economics and business school establishments

I read this yesterday, and he reminds me of Peter Schiff in that, he may be right about the problems, and he may have called them early, but he's clueless about the aftermath.

Marginalizing schools? Here's a crazy thought: how about make the schools more robust? We can't all be as naturally intuitive about black swans as Taleb; better instruction into the history of finance and economics could certainly enlighten people. Here's a link to one such paper (PDF).

One thing that shocked me about this entire crisis is how often these things have occurred in the last 70 years. And how similar their symptoms, and how similar the recoveries have been. And as I've discovered over the last 5-7 months (after reading a countless number of these papers), the information is out there. It's freely available. And casually ignored, because somehow "this time it's different".

Here's a pithy truism: The Black Swan is us.
posted by SeizeTheDay at 4:21 AM on April 9, 2009 [1 favorite]

Who needs black swan events when the Financial Times is posting Digg-bait 10 point lists? The Apocalypse is already here and it is being voted up.
posted by srboisvert at 4:44 AM on April 9, 2009

Don't they know that swans don't need money?

Speak for yourself, dude.

One thing that shocked me about this entire crisis is how often these things have occurred in the last 70 years.

The human race isn't very good about learning its lessons. The farther we get from the last correction, the more careless we get and the more likely to think things are different and that we're safe.
posted by orange swan at 4:47 AM on April 9, 2009 [4 favorites]

Thanks for the link to Klien jmhodges.

The cost of insuring black swan events is often excessive, just look to his hedge fund ventures. They made money, but given the scale of losses in the system, they did not make enough.

From the link:

Taleb spends a lot of time selling Taleb. This wasn't a Black Swan event. It was a bubble. A Black Swan is a meteor hitting the Alaskan pipeline, not an asset bubble popping.
posted by fistynuts at 5:02 AM on April 9, 2009 [1 favorite]

A big start in avoiding black swans would be to admit that prices of financial instruments are not (log)normally distributed or serially independent (or at least not dependably so).

Unfortunately, these assumptions are axiomatic in most risk management systems. Worse this critically undermines Black Scholes and much of the software that is used to not only derivatives but the monster ETFs that dominate some of the markets and result in massive end-of-day trades.

It's worth running your own tests if you have the capability to see just how not-normally-distributed and serially dependent most prices are.

It's possible to account for certain kinds of skew in Black Scholes but I think a lot of people are too lazy.

Taleb's best point in the article (to me) is

4. Do not let someone making an “incentive” bonus manage a nuclear plant – or your financial risks.

His point here is that bonuses reward risk-loving behavior. Let's say you have a trading strategy which produces a 50% annual return with 95% probability and has a 5% probability of a 1000% loss (this is a 10x leveraged strategy, quite mild by some standards).

The actual expected annual return from such a strategy is a big fat zero, but the manager makes a corking bonus nineteen years out of twenty, and gets fired once every two decades.

Now you add to this the fact that due to the non-normality and serial dependence of market prices, the '5% probability of a 1000% loss' is a serious underestimate, and you start to see how the whole thing works.
posted by unSane at 5:13 AM on April 9, 2009 [4 favorites]

We could just stop abusing that stupid metaphor.
posted by 0xdeadc0de at 5:20 AM on April 9, 2009 [1 favorite]

registration = FAIL
posted by gallagho at 5:24 AM on April 9, 2009

Of course Black-Scholes changed all that

One of the things I remember about Fooled by Randomness is Taleb's repeated assertion that Black-Scholes is junk and it "works until it doesn't," so is useless for mitigating the kind of risk he likes to talk about.
posted by adamdschneider at 5:36 AM on April 9, 2009

So let me get this straight, he says we should rebuild the hull of the school bus so we can make swan egg omelettes? The man is a metaphor machine.

Also:

Using leverage to cure the problems of too much leverage is not homeopathy, it is denial.

Because, overleveraged firms are administered doses of leverages so diluted that there is no actual leverage in them anymore?

I (still) know very little about economics and I don't really know whether he has a point, but I will say that IMO the writing doesn't help.
posted by goodnewsfortheinsane at 5:58 AM on April 9, 2009 [1 favorite]

Felix Salmon: "Derivatives are, at heart, bilateral contracts: how can you ban two consenting adults from entering in to such a contract?"
Steve Randy Waldman: "Contracts are not bilateral."
posted by preparat at 6:12 AM on April 9, 2009 [6 favorites]

Preparat: YES. That really helped solidify some thoughts I've been having.
posted by krilli at 6:27 AM on April 9, 2009

Taleb - taking better thinkers idea and claiming them as his own since 2001

I kid (sort of) - Fooled by Randomness should be required reading for anyone who works or desires to work in finance. Its just not really an original idea. Nor is he the iconoclast he claims to be.

Also isn't the concept of this article some sort of failure of logic? Black Swan - event you can't possibly rationally expect to occur.

Here is a list to make sure you avoid them.

Hmm
posted by JPD at 6:36 AM on April 9, 2009 [1 favorite]

JPD - Whose work?

And that's AFAIK not what the black swan idea means - it means that you have to be careful not to make assumptions that will instakill you 50 years on.
posted by krilli at 6:39 AM on April 9, 2009

Basically all of fooled by randomness bit on human beings being incredibly poor at thinking probabilistically is stolen from a ton of behavioral finance stuff. Most notably Kahneman and Tversky.

There are too many studies to mention on the inherent non-gaussian distribution of security prices.

And the concept of a "Black Swan" itself is Karl Popper's. Its from the famous story that Europeans used to say that something incredibly rare was "like finding a black swan" Then they found Australia and...well

And the bit about Wall Street thinking making money is skill, when usually its luck is pretty obvious to any one paying attention. Its also a big part of why everything blew up the way it did.
posted by JPD at 6:55 AM on April 9, 2009 [1 favorite]

Is this guy 17 or what?

In his late forties apparently.
posted by motty at 7:01 AM on April 9, 2009

Basically all of fooled by randomness bit on human beings being incredibly poor at thinking probabilistically is stolen from a ton of behavioral finance stuff. Most notably Kahneman and Tversky.

There are too many studies to mention on the inherent non-gaussian distribution of security prices.

And the concept of a "Black Swan" itself is Karl Popper's.

It's been a while since I read FBR, and I should reread it, but am I wrong in remembering that Taleb drops quite a few names in the book? I don't recall it being branded as, "These are all my original thoughts and have never been thought before."
posted by adamdschneider at 7:04 AM on April 9, 2009

adamschneider - yes the books are much less guilty of appropriating the work of others then his current spiel. But especially on the behavioral finance side he really doens't give enough credit where credit is due.
posted by JPD at 7:07 AM on April 9, 2009

FBR is a great book. Even if I think he presents his thoughts as more original then they are it is definitely something that everyone should read. I actually sort of find it uplifting.
posted by JPD at 7:10 AM on April 9, 2009 [1 favorite]

"Citizens must be protected from themselves." Because bankers and investors and government officials know what's best for you and hey wait a minute...
posted by eccnineten at 7:11 AM on April 9, 2009 [1 favorite]

adamschneider: One of the things I remember about Fooled by Randomness is Taleb's repeated assertion that Black-Scholes is junk and it "works until it doesn't," so is useless for mitigating the kind of risk he likes to talk about.

Back when I worked in finance the thing that struck me the most about all the crazy indicators and tools that people came up with was that they were all basically just trend following and rarely more informative than a triple moving average overlay. The well known techniques are universally junk.
posted by public at 7:14 AM on April 9, 2009

Mutant : What if we had an official route for determining if financial products had suitable models? To get a product approved, you need some consensus among interested mathematical finance people and a several million dollar bond held. If anyone can prove your model incorrect, then they take the bond home and you lose approval. You needn't ban financial instruments without approved models, but you must disclose their lacking when selling them.
posted by jeffburdges at 8:02 AM on April 9, 2009

JPD: Taleb gives a lot of credit to Kahneman and they appear to be friends. They appeared together here (video link). It's worth watching, if only for the Kahneman bits - Taleb's part are well... Taleb.
posted by pascal at 8:06 AM on April 9, 2009

Basically all of fooled by randomness bit on human beings being incredibly poor at thinking probabilistically is stolen from a ton of behavioral finance stuff. Most notably Kahneman and Tversky.

I don't think you can say that Taleb "stole" Kahneman and Tversky's ideas in FBR, since he attributes their discoveries them by name dropping those two on practically every page in the book, and even frequently refers to Kahneman as "Danny." Hardly the M.O. of a plagiarist and idea thief.

Maybe his presentation of their psychological research isn't original, but it's a far shot from a blatant rip off.
posted by Hello, Revelers! I am Captain Lavender! at 8:14 AM on April 9, 2009 [1 favorite]

Protecting against all risks, especially catastrophic risks like Black Swan risks, would be incredibly inefficient and expensive

That would be a terrible idea.
posted by Pants! at 8:31 AM on April 9, 2009

Stolen was a poor choice of words. I realized that after I wrote it.

My issue really is the media who seem to present Taleb as if he were the originator of his ideas.

Also I find him irritating.

Am enjoying that video as we speak tho.
posted by JPD at 8:40 AM on April 9, 2009 [1 favorite]

@ 9:30 of that linked video Taleb acknowledges that he had to rewrite FBR after he met Kahneman. I read the first edition. I wonder if that's why I had a different impression of his willingness to acknowledge the work of those before him.
posted by JPD at 8:46 AM on April 9, 2009

Protecting against all risks, especially catastrophic risks like Black Swan risks, would be incredibly inefficient and expensive

That would be a terrible idea.

To me, that's not the question. Rather, this is what happened:

We willfully ignored all risks, especially catastrophic risks like Black Swan risks, it was incredibly inefficient and expensive

It was a terrible idea.

The two statements are not the negation of each other.
posted by krilli at 9:24 AM on April 9, 2009

I really hated The Black Swan. It made my heard hurt trying to figure out what he was saying, and I am quite familiar with Kahneman and Tversky's work, and non-EU decision theory in general.

If his point is that human attempts to categorize real-world processes are reductive, and that categorization is a fragile exercise in the face of real-world change, I think he makes an excellent point. If he is claiming this problem is compounded seriously by systematic mistakes people make assessing probabilities and weighing probabalistic outcomes, he makes another good point. But it's never so simply or clearly stated, and these points are not new. He really grossly mischaracterizes the state of research in economics and history, and I assume other fields as well.

There's a passage in TBS in which he describes giving taxi drivers $100 tips, essentially describing doing so as an unexpected and wonderfully transformative moment for the taxi driver. Anyone with this degree of megalomania and utter lack of introspection scares the hell out of me.
posted by ~ at 10:23 AM on April 9, 2009

I don't recall it being branded as, "These are all my original thoughts and have never been thought before."

Though funnily enough, I actually do know someone who used to work with him who claims that some of his (guy I know's) ideas were used without attribution. Can't prove it by me, but the guy is a pretty straight shooter.
posted by IndigoJones at 11:01 AM on April 9, 2009

How have we not linked to his notebook yet? It is a much more interesting and illuminating read than his popular writing, and it will quickly alleviate any suspicion that he is claiming anything as his own idea (he is doing quite the opposite, attributing almost everything he says to one ancient or another). I'd recommend reading from the bottom up as his recent entries are suffering from an ego-trip.
posted by TwelveTwo at 11:08 AM on April 9, 2009

Mutant : What if we had an official route for determining if financial products had suitable models? To get a product approved, you need some consensus among interested mathematical finance people and a several million dollar bond held. If anyone can prove your model incorrect, then they take the bond home and you lose approval. You needn't ban financial instruments without approved models, but you must disclose their lacking when selling them.

Jeffburdges, this is an interesting thought. If economics were an actual science, with the checks and balances of things like peer review and institutional ethics, this might work.

Sadly the entire field is hamstrung by two inescapable realities:

(a) Despite protestation by economists about economic "laws", it is NOT the study of the natural world with laws akin to the laws of chemistry, physics, biology, etc. It is a study of a human system limited only by the boundaries of human imagination and human intent. In essence, studies in economics are all premised on "What happens if we agree to pretend X, Y, and Z". Hard to prove a model incorrect because there are no physical realities to provide a reality check.

(b) Most work in the field is shaped by pursuit of profit. Conflict of interest is almost inevitable and the integrity of the field is inevitably compromised.

Love the idea from the list of getting rid of the Nobel in Economics. God, what a waste.
posted by Sublimity at 11:25 AM on April 9, 2009 [1 favorite]

Also I find him irritating.

He does indeed have no shortage of self-regard.
posted by adamdschneider at 12:59 PM on April 9, 2009

An alternative take by Amartya Sen: Capitalism Beyond the Crisis.
posted by russilwvong at 2:32 PM on April 9, 2009

Protesters crash Summers talk
posted by homunculus at 2:50 PM on April 9, 2009

Of course Black Scholes changed all that

He's like the Rio Ferdinand of finance?
posted by mr.marx at 2:56 PM on April 9, 2009

I've read both books and watching some of the comments I can tell that most of the people who are criticizing Taleb haven't done that. Both books are stock full of notes, references and acknowledgments. FBR has 43 pages of such. (I don't have TBS handy to check at the moment). He also makes reference to his website, which contains even more source material. Taleb doesn't indicate that he's the first to think of these things, but he does seem to be the first to distill these various sources into one place and utilize them in a comprehensive analysis of the world. Essentially a scientific method that isn't reductionist, but at the same time provides useful information to make decisions.

TBS is a hard book to simply sit and read. Once you're able to grasp what he's talking about though, it's the sort of book that changes your view of the world. It also takes on a difficult subject not seen in other non-fiction works available on your local Barnes and Noble bookshelf. If people know of others, I'd love to read them.

What Taleb has done successfully with TBS is link Tversky, Kahneman, Popper, Mandlebrot, and others in a way I have yet to see anyone else do. He's elucidated and pulled together concepts from separate "silos" or disciplines and pointed to a need for a new framework for learning about the world.

What he seems to be attempting to do currently is develop that foundation on which to build knowledge. These rules let you make decisions in the face of uncertainty, essentially, make decisions in the real world. Although we currently make those decisions every day (since we have to live in the world), we do so by ignoring the possibility of disaster. The money shot is having a system that allows decision making that doesn't ignore that possibility, but rather minimizes it's impact (ie: avoiding death, financial ruin, etc.).
posted by herda05 at 3:23 PM on April 9, 2009

I don't think he's talking about stock options. We've got housewives trading these things today...

I know this was just a throwaway comment, but please don't.
posted by birdie birdington at 7:06 PM on April 9, 2009

The title of this should be "Ten principles for an economic Black Swan-proof world". There are many other systems subject to rare, game-changing events than just economics. If we're to create a truly resilient society we need to start addressing those systems as well.
posted by scalefree at 2:53 AM on April 10, 2009