Nassim Taleb (Black Swan) open letter to David Cameron
guardian.co.uk
guardian.co.uk
"David, you must counter this complexity by lowering indebtedness. We have known since Babylonian times that debt is treacherous and allows no room for mistakes: felix qui nihil debet goes the Roman proverb ("happy is he who owes nothing")."
He sounds like Ahmedinejan in his letter to former president Bush advising him to embrace the ways of the prophets of old :)
There are many situations where debt is your friend but like all good things it can be abused. Common Taleb, find something new to talk about! Black Swan is so 2002!
Secondly, regarding insight. This idea which everyone credits to Taleb is of course described all over the place. For example, as "the impossibility of inductive reasoning in the natural sciences" in The Mathematical Experience. As far as fat tails are concerned, every textbook on financial maths I own talks about fat tails the corrections that must be made to account for them, and they were all published before either of Taleb's books.
Absolutely not. Ideas that are diffuse and unoriginal, on the other hand, are often repeated because they are uncompelling.
I've read The Black Swan, and although it was entertaining, I didn't feel like I learned anything of value.
Bastiat articulated the Broken Window Fallacy in 1850, and here we are 159 years later shredding perfectly good cars to 'create jobs.' It certainly seems like some good ideas can't be repeated often enough.
I agree that commercial debt can be used for investment and is therefore useful. I think he is commenting on national debt. The problem is that the UK and US hold a magnitude of debt that will not be re-paid for several generations. Most of us would agree this is not healthy.
Further to this, his point about debt is that it is used to prevent the old institutions from failing and the resulting tax increases are hindering the next wave entrepreneurs. The financial crisis should have sped up the next wave of creative/destructive innovation in the financial sector but instead we are pretty much back to the status quo in The City and on Wall Street.
"Then we will see an economic life closer to our biological environment: smaller companies, richer ecology, no leverage. A world in which entrepreneurs, not bankers, take the risks and companies are born and die every day without making the news."
Having been on both sides of the coin, I do wonder how much those who allocate the capital (bankers) should make relative to those who generate the ideas and innovation - but clearly in recent years, it was skewed towards the former. Lots more from Taleb at his site: http://www.fooledbyrandomness.com/
He sees current regulation as helping to create the monsters we have now, where the risk-takers are allowed to play with the money that's supposed to be stable.
But how do you get there? How do you make sure they don't spin out of control or even how do you decide which firms to break up and how? (I don't disagree, I am genuinely interested in learning how he sees it can be implemented and without new significant regulations no less).
But if regulators would forbid any product valued by models they don't understand, it would probably be a good start.
Mocking the young math-geek (I studied math) analysts, bankers, derivative creators or whatever their titles are, is a good start. I heard it somewhere that over the past decade or two, everyone's mom went from wanting their kid to be a doctor or lawyer to instead be a Wall Street I-banker. All we really need is for society to trust elaborate investment and debt schemes less.
I'm not sure if the moms had it right the first time though, because we have too many lawyers and doctors are taking a beating now from Obama,
He could have been saying "You and your party may be the only hope we have for a resilient society insulated from negative oranges and in which everyone has the opportunity to benefit from positive oranges." for all the difference it makes to me.
Black swans on the other hand are birds with a particular pigment.
Now what are the additional layers of meaning, we see black swans referring to apparently unlikely events that nevertheless will occur (that's possibly not the definition given by this guy), but what about the super-definitions for the others?
(and even though I did start reading his book and do sort of know what he's on about I still cant get away from that feeling. Why cant he say "lets not be arrogant and high risk" without having to qualify it with IMO a poor reason (surely the sentence alone should be enough):(
I'm not sure he'd agree with you. Taleb's books repeatedly stress the importance of survivorship bias. In any population, they'll be someone forecasting just about any conceivable event. No matter how unlikely events turn out to be, somebody is eventually going to be right, and then that person is lauded as a genius for having forecasted correctly. When really they've done no such thing; rather, people have just retrospectively picked out the person who happened to guess right.
I think Taleb would say you shouldn't dismiss him because there's ample evidence throughout history that he's right, not because he happened to guess right this time. Whenever debt levels have risen to what they are now, it's resulted in economic instability and chaotic behavior. It's not just this century: you can find examples back to Roman times.
A Goldman Sachs executive just boasted that their business methods hadn't changed at all from before the crash. Think about that for a second.
From a financial viewpoint, the markets exist to transfer capital from those that have it to those that need it.
No, that's what communism was supposed to do. Under capitalism, we use the market to allocate capital efficiently, to its most productive use. Perverse incentives, like excessive bonuses or government bailouts without penalties, interfere with the efficient allocation of capital.
As for some god-given right to bank accounts, I never said any such thing. It's not my fault, nor Wall Street's fault, that people aren't better prepared to understand finance. It's a dog eat dog world. If you want to keep hold of your money, learn how to invest for yourself, place it with a broker and accept that even they may lose money, or hide it in your bed. Those are your options.
Under capitalism, we use the market to allocate capital efficiently, to its most productive use.
You are not distilling finance enough. Your definition is a first-year economics definition. That's a definition from an idealized world. In the real world, companies exchange shares for cash. They invest that cash into the business to create value. Other instruments have their own purposes, but it's all about taking either money or risk and transferring it from one party to another based on needs.
But do we have no middle ground here?
We agree, for instance, that most people really AREN'T equipped to understand finance as it is practiced. The difference is that you believe there is a class of people whose advice can be relied upon. Taleb thinks not for epistemological reasons. The perverse incentives I mentioned are (for him) barely worth discussing, since they are just corruption rather than self-delusion.
(And when I say corruption, I don't mean they are necessarily evil people; I mean their agenda is now misaligned with the customers').
Anyway, I keep saying "Taleb says" because I am not well versed in these matters. I study it about as much as a layperson can, and all I know is that most of it still seems too complex to predict.
As for finance being complex, it's really not.
look, by no means am i blaming 'widows and orphans' for this mess, it's just the reasons for it are very complex. even regulation that seems like a great idea, grading investment products, ended up playing some small part in the chaos.
http://en.wikipedia.org/wiki/File:USDebt.png http://en.wikipedia.org/wiki/United_States_public_debt
Margaret Thatcher when asked what was her greatest political achivement - 'Tony Blair' she replied. Isn't Cameron just a more greasy version of the slimy Blair, Black Swan spotter or not?
I stopped reading here. The only objective evaluation of economic matters is in scientific, risk-and-return terms. Anything else is quackery, fear-mongering, or worse.
A great deal of evil has been done in the last decade or so in the name of protecting us against horrible things. We are told that much good has come of it, because nothing horrible has happened. While I'm sure this is partly true, I wouldn't mind trading a few more horrible things for a little more autonomy.
That's not the problem; I suggest you read up on his theories. His point is, to put it more mathematically, that using inappropriate statistical models to try to model future events can lead to catastrophe. In particular, assuming something is a normal Gaussian process when in fact it's something that's almost a normal Gaussian but has a slight propensity towards total disasters that occur more often (and much stronger) than the Gaussian distribution would imply can result in you creating a system that is terribly brittle when those "black swans" occur.
"Scientifically" applying the Gaussian distribution or other advanced mathematical constructs when they manifestly do not apply is the real quackery, if Taleb is right. And it's hard to look at the evidence and conclude he's wrong here; it's simply objective fact that many of the interesting distributions are in fact prone to "black swans", there's a statistical test that shows it (kurtosis) quite clearly. It's really more a question of what we are going to do about his point than whether or not he's correct about the distribution. (There's no guarantee that his prescription is correct, but it's probably a better idea than listening to the "experts" who advise you to mortgage to the hilt because the worst case scenario can't possibly happen and you are 99.9% percently likely to be fine. You know, that worst case scenario that just happened and we're still living through. Yeah, that scenario that can't possibly happen.)
He says: "We replaced the heuristics of the elders with arrogant (and incompetent) beliefs, breaking, in the name of science, the chain of knowledge."
In europe we had fascisim and stalinism within the past 100 years. That's three or four generations. Before the dawn of liberalism, capitalism, enlightenment and the industrial revolution we had 1000 years of depression caused by religious fanatics. We had to break that chain of "knowledge".
The heuristics of the elders are little more than mature, cultivated cliches of what past experts and ideologues told them. Taleb is correct, it's incredibly difficult to make predictions in a complex system. However, I don't see how a blanket condemnation of science and a blanket approval of crude traditions helps us to become better at that.
Taleb's position (and argument in this paper) is that we need to recognize instances where we don't know the answer instead of making decisions based on assumptions that we do.
The housing bubble was BLATANT. No black swans involved. I remember talking about it with friends in 2005. No big deal, really.
Yes, but what was not blatant was that the housing bubble popping would kill Lehman Brothers, cause runs on banks and almost bring down the whole world economy.
Taleb's point is that you can't go around behaving like any system is, to borrow a phrase "too big to fail" - or rather too cleverly thought out/regulated to fail. To some extent this is simple engineering. I can test the resilience of my cluster to network failure by pulling out an ethernet cable and see what happens - maybe there's a small hiccup while systems fall back onto local resources, maybe nothing fails over properly and I am hosed for 12 hours. As I understand Taleb, his argument is that the economic powers (bankers, traders, politicans) spend too much time trying to figure out how to prevent network failure, rather than making sure that when the network does go out, it doesn't take your whole mission down with it. Because, no matter how well you think you have engineered the system, the chance of your network going out somehow, sometime is non-zero.
I do agree with the posters that Taleb overstates the cleverness of his own insight; but unfortunately, in his field, it seems that people really were too dumb/greedy to grasp this. Irrespective of what you think of his interpretation of events or the validity of his suggestions (and I too question them), I think as systems engineers we can appreciate the risk of designing political and economic systems under the assumption that they can't fail.
But Taleb makes a different point: Discounting the apocalypse is exactly what he's arguing against, because this is essentially what LTCM et al were doing with their models. They calculated probabilities to make sure that the estimated payoff was greater than the estimated loss, but they got the distributions wrong. What you should try to do is make sure that the apocalypse can't happen, that you aren't exposed so that one outlier will take you down. Because due to the nature of the system (there is only a sample of one of the world economy) you simply can't gather data that can pin down the wings of the distributions to the precision you need.
According to the models used at the time, the Black Monday 1987 was something like a 30-sigma event. If the distributions truly were gaussian, such an event should never happen in the history of the Universe. So you have a choice between thinking that we just had an incredibly bad luck, or the model is wrong. I'd go with the latter.
Some data, some math, and an interpretation you won't hear every day:
http://quickfacts.census.gov/qfd/states/00000.html
http://photos1.blogger.com/photoInclude/x/blogger/2825/754/1...
There are 127 million housing units in the U.S. At the peak of the bubble, 7 million houses/year were sold. A total of roughly 30 million houses were sold in the boom years of 2002-2007, assuming there's no double-counting (people who sold a house multiple times), which is probably a pessimistic assumption. That means that over half of U.S. households took absolutely no part in the bubble - they just sat tight.
Now look at the inventory numbers from the last two years. They go up rather dramatically. This means there were more sellers than buyers, i.e. a large number of people realized that home prices were selling for more than they thought the house was worth, and quickly put their house on the market to take advantage of this. Sure looks like people recognized the bubble to me, it's just that by definition they can't all get out before it bursts.
My interpretation is that most people did recognize the bubble - it's just that they're not the ones that the New York Times writes stories about. They held onto their home and neither bought nor sold, instead just going about their business. Or they sold at the peak, rented for a bit, and are now buying back in at half price and pocketing a few hundred grand. Smart people tend not to brag about how they just made a killing off other people's stupidity. It prevents people from being stupid again, which limits the opportunities for future killings. Besides, it tends to kill the mood at parties.
http://finance.yahoo.com/q/bc?s=VEIEX&t=5y
...I wouldn't have made any connection to US housing and foreign equities, but I was looking at both, and I knew both were bubbly. I've since lost money on a few stupid trades, so it's not like I'm an expert or anything. But the fundamentals are usually clear to all. Right now, the macros on S&P PEs, long-term treasuries, and high-yield,
http://finance.yahoo.com/q/bc?s=JNK&t=2y
...are clear, and they all scream "avoid". And that's the extent of my trading strategy right now.
Have you ever worked at an investment bank? I suspect you have not. Wall Street does not run on the scientific method. It runs on greed, fear and politics. The ones who check the assumptions and know the limits of their models don't have the power to make the decisions. This is the problem. And the Taleban apparently haven't yet figured it out. Quoting Janet Tavakoli: "malfeasance, not models, disrupted the global economy, and Taleb still gets that part wrong."
Last but not least: there's not such thing as a "blackswan theory". You can't steal ideas from probability theory and statistics that have been known for centuries, rename them, wrap them up nicely, and call them new or original.
Taleb is an anti-intellectual. He believes that all models are wrong and useless. To make it worse, he advocates that no further knowledge can be attained. The only knowledge and truth that is right is the "black swan theory". Come on, gimme a break!!! Aren't we HN'ers smarter than that? Why is Taleb's drivel being upvoted?!
Do you have any references to work centuries old that describes what Taleb calls 'Black Swans'?
Last but not least: in the late 1990s, Lehman Brothers had a trading desk dedicated solely to extremely unlikely events. That was before the so-called "black swan" theory came along. They made money only once. On 9/11.
Taleb's good ideas are not original, and his original ideas are no good.
There is a lot going on in finance every day that simply cannot be objectively modeled or understood in scientific terms. That's a problem, but we also need to accept that there may not be a globally valid solution. Viewing complexity as the enemy sounds like as good an approach as any.