Felix Salmon

Taleb’s necessary and impossible wish-list

By Felix Salmon
April 8, 2009

I hate listicles. But every so often, one comes along which is utter genius, and Nassim Taleb’s list of “Ten principles for a Black Swan-proof world” is one of those cases.

Yves Smith calls it a must read, and one which she is “highly confident will never be implemented”; she’s right on both counts. But in many ways that’s the strength of this piece: it both must be implemented and can’t be implemented at the same time. Which constitutes a massive policy dilemma.

Taleb’s first principle is that “nothing should ever become too big to fail”. But all economies have too-big-to-fail institutions; they always have, and they always will. Looking at the rest of the list, how on earth do you stop the financial sector from awarding its employees bonuses, or creating complex products? Derivatives are, at heart, bilateral contracts: how can you ban two consenting adults from entering in to such a contract?

Taleb’s big idea, these days, is that we of necessity are moving into a world with vastly less debt than we’ve been used to. I’m not entirely clear on what his timeframe for this is. A quote from my (boss’s boss’s) boss, Tom Glocer, seems germane here:

“I’ve met a lot of smart people in my life, and they’re the ones who are eventually always right and they always know where things are going [and] they always underestimate friction in the world and how long it takes to get there.”

22 comments so far | RSS Comments RSS

A few years ago there was an excellent TV programme in the BBC’s Horizon series (you guys call it Nova). Ostensibly about the history of bridge design, it was a actually about the oscillation found in many human activities between a cautious, over-engineered, expensive phase, which leads gradually to a highly optimised, graceful phase, which after the next big disaster leads back to the first phase.

You can’t stop this cycle, and you probably shouldn’t even want to.

Posted by Ian Kemmish | Report as abusive

Well, I am not sure I agree with this: Taleb’s what I call ‘April Theses’ have been put forward by him a number of times before. What strikes me is the sheer lack of specificity and practicality – this is not a massive policy dilemma, this is simply an indication of the limits to how much we should be listening to Taleb.

What I find most interesting is his thoughts with regard to complexity. On a recent EconLib podcast with Russ Roberts, Taleb made the points the globalisation increases complexity thereby leading to systemic fragility. The solution therefore is to simplify, including but not limited to financial markets. Paul Wilmott make a similar point today on his blog when he takes about society reaching the limits of its own incompetence. This is a very powerful observation, I think: modern life has become so complex as to challenge our abilities to cope and our competence.


Taleb is trying to sell his book. Taleb’s thesis is not novel: Mandelbrot-Levy distributions (‘black swan’ or ‘fat tail’ curves) are indeed nearly impossible to predict. However, his remedy of basically giving up is but a modern spin on David Hume’s skepticism, and is not practical. Better to simply be a bit less greedy and more fearful, and to abolish “too big to fail”, which encourages greed.

Posted by Ray Lopez | Report as abusive

While on the surface many of Taleb’s principles seem logical, most of them are nonsensical and impractical. The article is full of clever sound bites and fallacious analogies, but very short on anything of practical value.
1. Most institutions are not fragile when they are small, but become fragile by growing too large, too fast. We will always have too big to fail institutions.
2. Nationalizing financial institutions is a bad idea the government is not capable of running a business. Saying that the banks took over the government makes a nice sound bite, but it has no basis in fact.
3. Agree
4. It depends on how the incentive bonus is structured, but I would tend to agree with his statement.
5. Statement makes no sense, complexity on its own is not a problem- complex instruments in the hands of the ignorant are dangerous. How does this statement translate into action?
6. A very naïve statement. While there have been many derivatives that have been abused, there are many that are useful. What we need is transparency and regulation, not an outright ban.
7. Too ridiculous to comment about.
8. Okay, and…
9. ???? Is he saying that all citizens own their own business? How are people supposed to save for retirement?
10. Too ridiculous to comment about.



I’ve tended not to be a fan of pension plans because they tie an individual to a company and expose them heavily to that company’s credit risk. I’ve become a bigger fan of pension plans for financial executives in the past several months.

Confidence is, like it or not, an important part of any modern economy, and most primitive ones, even. Tyler Cowen once suggested a sushi indicator: when sushi (sold on the impersonal market) starts to become popular in a location that previously had little sushi or market trading, that economy is ready to become “developed”. We place a certain amount of trust in strangers (and/or “the market”) just going grocery shopping, particularly when buying something like fish that we expect to eat raw, and the levels of division of labor necessary for that require confidence.

A different kind of confidence — confidence in the future — is necessary for any capital investment decision. Installed capital is worth only the NPV of what its use will be worth in the future; if people are only willing to accumulate capital when it’s likely to repay itself in 5 years or less, you’re going to have a low-productivity level of capital. Capital may not be a great way to store value, but it’s by far the best way possible.


When Taleb says “Whatever may need to be bailed out should be nationalised,” he doesn’t believe the government should run it. He means it should be taken over and liquidated in an orderly fashion.

Posted by Dave | Report as abusive

> “Taleb’s first principle is that ‘nothing should ever become too big to fail’. But all economies have too-big-to-fail institutions; they always have, and they always will.”

The problem is not that an organization is “too big to fail”; rather, the problem is that the state mistakenly regards organizations as such–which leads to bailouts (i.e., wealth extorted from the productive and funneled to those failures deemed “too big”).

Thus, “too big to fail” is an erroneous category. It is better if a enterprise fails without receiving taxpayer-extorted funding; otherwise, the moral hazard is an incentive to larger failure in the future.


Debt levels fell gradually for over a decade after 1929. It won’t happen overnight and even as it happens it won’t prevent the economy from growing.

Posted by Lord | Report as abusive

just my 2 cents after reading taleb’s book: i really believe this guy is hugely overrated. i mean, the big idea was what: that there are instances when extremely low probability events might happen? and it gives you a headache to read it whether you are an academic or a layperson.
in this case the computing the probability was the problem.

Posted by camelback | Report as abusive

“all economies have too-big-to-fail institutions; they always have, and they always will”

Felix, have you heard of the Great Depression? No too big to fail institutions there. Reminds of court cases where pleaders always claim that X was the custom “since time immemorial”. It was never true, but you could always hope the judge believed you.

Posted by Anon | Report as abusive

Like much of what Taleb has written since “Fooled by Randomness”, this piece is best understood as an intellectual tantrum. It’s not meant to be implemented; it’s meant to take yet another set of swipes at those who “didn’t listen” to NNT back in the day. Isn’t this guy’s 15 minutes up yet?

Posted by John | Report as abusive

Derivatives are, at heart, bilateral contracts: how can you ban two consenting adults from entering in to such a contract?
The same could be said of securities, and they are heavily regulated.

Posted by Pender | Report as abusive

1) Breaking up companies too big to fail is what antitrust law was invented for.

2) So you’re saying we can’t ban agreements between two consenting adults, like prostitution and heroin sales? Governments control people all the time in this way. The question is not “can we?” but “should we?”

Posted by Eric | Report as abusive

I went for a railcar excursion last summer. Railcars are those wee tiny railroad maintenance vehicles, smaller even than an automobile. They STRONGLY emphasized safety, and pointed out that multiple safety measures exist because things fail. Sometimes multiple things fail. And when they all fail at the same time, you get what Taleb calls a Black Swan. The key to eliminating most Black Swans is to have safety procedures in place, and follow them every time.

Railroaders have a rule (actually a rulebook, and their luggage even has a special compartment for holding it — because they MUST have it with them when operating a train): any train, any track, any time. ALWAYS look both ways and listen before you cross a track. NEVER step on a rail (they can be slippery).

Most railroad Black Swans don’t happen, which makes me question why Taleb worries so much about Black Swans. Maybe he’s just trying to avoid having his book becoming forgotten?


I have read Taleb’s list several times, and I find much of it really obtuse, but this one takes the cake:

“9. Citizens should not depend on financial assets or fallible “expert” advice for their retirement. Economic life should be definancialised. We should learn not to use markets as storehouses of value: they do not harbour the certainties that normal citizens require. Citizens should experience anxiety about their own businesses (which they control), not their investments (which they do not control).”

What exactly does this mean? That we should save for retirement only by purchasing physical assets? That we should all own our own businesses? That all financial assets, including money are snares and delusions? That we should not retire, but should work until we die? What does “definancializing” economic life mean? Reducing the size/importance of the financial sector? Eliminating financial instruments?

Making vague, even gnostic, pronouncements may gain you a reputation for wisdom (the obscure are often thought wise, when they are mostly only obscure). But I defy anyone to explain how we would do all (or, perhaps, any) of these 10 things.

Posted by Donald A. Coffin | Report as abusive

Taleb’s claim regarding derivatives is childish and ignorant. Just because he doesn’t understand them doesn’t mean no one does. They did not create themselves. Derivatives are essential to modern economies, and have been for hundreds of years. Poor regulation, primarily of banks’ risk taking but also of derivatives markets, are to blame, not the instruments themselves.

Posted by Robert Williams | Report as abusive

“Taleb’s claim regarding derivatives is childish and ignorant. Just because he doesn’t understand them doesn’t mean no one does.”

Robert, I have an inkling he does understand them.

Jobs he has held:
“Managing director and proprietary trader at UBS. Worldwide chief proprietary arbitrage derivatives trader for currencies, commodities and non-dollar fixed income at CS-First Boston. Chief currency derivatives trader for Banque Indosuez. Managing Director and worldwide head of financial option arbitrage at CIBC-Wood Gundy. Derivatives arbitrage trader at Bankers Trust, proprietary trader at BNP Paribas, as well as independent option market maker on the Chicago Mercantile Exchange. Founder of Empirica LLC” (Wikipedia)

Posted by Dismalist | Report as abusive

Derivatives can be explained in simple English. Just because very few people have the desire or the talent to do so does not make derivatives by themselves too complicated to be understood by regular folks.

If you hear someone say this or that financial product needs to be kept away from the great unwashed because it’s too complicated, hide your wallet.

Posted by vachon | Report as abusive

I think Taleb’s theory and this list both make sense. I think some of the forum commentators view it too literally or too generally. Ask any sound businessman – he will tell you the secret to great business/management is simplicity and transparency (predictability).
Complex products are built on numerous variables some of which have been approximated, tailored and assumed for the sake of convenience – they work fine as long as the inherent variables are predictable. In an interlinked world economy, the interplay of related and unrelated factors impact the behavior of those inherent variables in ways which can not be predicted even by ‘experts’ – including their timing. In that sense, they defy the understanding of even those who created those products and/or those who use the products. It may therefore be appropriate to ban them.
Nationalization could be part of solution at this time – it works in the emerging countries. Agreed they are slow but they are also steady.
I disagree with Felix on policy dilemma. There is none. It just has to be implemented. Right – time frame is not mentioned. I would settle for a beginning and a small steady pace. Hopefully inertia will be overcome after a few first steps.
I think the biggest challenge for regulators and businesses is to manage the alignment of objectives, interests, incentives and horizon (time frames) between the customers and market participants.

Posted by Satish | Report as abusive

@Robert Williams/vachon:
Taleb’s hedgefund made millions with trading derivatives on the assumption that humans make certain errors with valuing derivatives the wrong way.

Taleb doesn’t argue that derivates are to difficult to be understand by regular folks. He argues that they are to difficult to be understand be the bankers and that because those bankers misunderstood them they crashed their companies.

@Forbidding Derivatives:
The US also forbids prostitution with is a lot less harmful to society.

“Saying that the banks took over the government makes a nice sound bite, but it has no basis in fact.”
Geitner was beforehand head of the New York Fed. Six of nine votes that control the New York Fed are in the control of banks. Those New York banks thought that Geitner was the best person to represent their interests as head of the New York Fed.
Now Geitner is treasury secretary. Why do you think that Geitner and co haven’t taken over the government?

Posted by Christian | Report as abusive

I just made a posting at my blog (http://texasholdeminvesting.com/2009/04  /nassim-talebs-10-rules-to-black-swan-p roof-the-world-my-rule-11-use-texas-hold em-poker-to-learn-to-invest/) about this article.

While in one sense I agree with Nassim, in three areas I don’t:

- There are some parts of the modern capitalist society that required large companies or entities to actually work e.g. try building silicon wafers in your back shed. You can’t de-risk this type of situation completely.

- The economic environment can never match the biological one. The economic environment is completely at the mercy of human psyche and emotions, which don’t operate to scientific principles like the biological world.

- I’m not sure I want the economic (or social) world to resemble the biological world. The biological world can be a frightening world – species are exterminated by other species with humankind’s help, the law of the jungle prevails, and environmental catastrophes can and do happen.


Diving deeper into the contents of Taleb’s Ten Principles, I believe that the rot goes much deeper into a “diseased” Western Civilization that transgresses any geographical boundaries. I wrote an article recently on the Derailment of Western Civilization- Does it need a bail-out? It may be worth a read in context of the diabolical intent of guys like Larry Summers and Tim Geithner at the helm of affairs. http://journals.copperstrings.com/UserCo nsole/ViewJournal.aspx?Title=The_Derailm ent_of_Western_Civilization_-_Does_it_ne ed_a_bail-out%3F&ArticleID=1304


Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/