In an earlier post we shared some thoughts about robustness in interconnected networks. In 2010, Nassim Nicholas Taleb held a presentation for the Henry Jackson Society. Taleb discussed the global financial crisis and his Black Swan Theory, elaborating on his enhanced second edition of The Black Swan which expands on his views as to how to cope with improbable and potentially devastating events, not through forecast, prediction and reliance on statistical data, but by adhering to the concept of ‘robustness’ in the face of potentially devastating systemic fragility.
The problem with technology is that when it works, it is better; when it does not work, it is far worse. On one side, we have something very robust that does not appear to be efficient. On the other side, something that is seen as efficient, technologically developed with perceived benefits, but it is much more error prone; when it breaks, it breaks. There was a blackout in New York in 2003. I went to the bathroom on the 32nd floor and I could not wash my hands because there was no water; the water was dispersed by electronic hand sensors. A faucet may be less efficient, it may be less hygienic, but it works all the time.
In a nutshell, this represents my problem with statistics. You can only say something is efficient, after it proves to be so, not before. In other words, gains in efficiency only count at the end of the gain, not at the beginning.
So let me rephrase my idea of the Black Swan to make it less cryptic. I would like to discuss the idea of the Black Swan, before discussing robustness and the characteristics that make a society robust: a robust eco system; a robust economic system; and a robust banking system.
I wrote the first 350 pages of the Black Swan before the crisis of 2008 and it is about a class of events that you do not see coming that can hit you very hard. I borrowed the metaphor from the Roman poet Juvenal whose phrase was used in London in the 16th century, to say that something was “as rare as a black swan” because there were no Black Swans in the old world. My Black Swan is an event, not an exception or a bird with feathers, but an event that has big impact when it happens. But it has another characteristic that some people could not understand when I finished the first edition of the Black Swan and that is that a Black Swan event for the turkey is not a Black Swan for the butcher. When people asked me if 9/11 was a Black Swan event I would tell them that it depends if you were on the plane as a terrorist or a civilian in one of the towers. For the terrorist it was not a Black Swan.
The second thing I described in the book is that there is an environment that creates Black Swan problems; a man made environment that I call ‘extremistan’. Ancient environments did not have Black Swans, but the new one does and it comes from several features including technology, winner takes all mentalities and so on. We live in a society that is very different from our ancestry one. So, economic theory becomes very hard to forecast because you can not forecast Black Swans, you can not forecast history in ‘extremistan’.
If you gather 1000 people and add to them the heaviest person on the planet, that person would change the average weight very little. That person would represent no more than 0.3% of the total and if you move from 1000 to 10,000, then that person will be 0.03% of the total. So the exception will not play a large role in an environment I call ‘mediocristan’. Everything in economics is based on that concept; that mild domain called ‘mediocristan’. Now if we play the same experiment with the same 1000 people and you add the wealthiest person to the sample, that person will be near 100% of the wealth; £50 billion versus an average of probably less than £1 million. So they are actually in an environment, ‘extremistan’, where the exception plays a very large role and where a Black Swan can really dominate all the properties.
In my theory prior to the crisis, I discussed fragility to Black Swans within the banking system where you had one single incident bringing down the entire system because it was very fragile and because the economic establishment were using the wrong statistical tools.
The UK is a much more robust country than the US in the sense that it recognises that the economic models and tools that we are using do not work whereas in the US the economic establishment still believes in these models.
The Office of Management and Budget is producing forecasts on which the Obama administration is basing its deficits, but they have not looked at the forecasts from the past. Similarly, I was at a conference in Korea and a gentleman from the IMF was showing us forecasts for 2010, 2011, 2012, 2013 and 2014 without showing us his forecasts for 2007 or 2008 or even 2001, 2002 or 2003 and so on. I do not want to change human nature; I just want to change the system wherein these people can be incompetent without being accountable.
And there is a moral point here too. When you have adults making mistakes, say for example one individual who does not understand risk lending to another individual who does not understand risk and both miss-forecast, then who pays the price? When you print money – quantitative easing – people doing the right thing pay for those who make mistakes. And what is immoral here is that your children are going to pay the price. That is a violation of the old Roman adage that you do not pay for your ancestor’s mistakes.
After the crisis, people asked me what we should be doing. The logical conclusion is to stay as far away as possible from certain exposures that make you Black Swan prone.
The first is debt. If you have low debt in society, you are much more resistant to Black Swans. For example, consider two brothers; each one of them has £100 of capital and forecasts predict that their respective projects will bring 6% of returns. The first one issues equity and shares in contrast to the second one who borrows as much as he can. The one who borrows will have much greater returns but will be much more error prone and will need to be very precise in his forecast for the future. One single error in forecasting and he is bust; he is gone. Debt makes you a lot more fragile to Black Swans.
We can learn from Mother Nature which is designed to be robust to Black Swans. It has reserves; the exact opposite of debt. If I had £100 and invest all of it, I do not have reserves and am therefore prone to Black Swans. If I borrow an extra £100, I am even more prone to forecast errors and by extension, more prone to Black Swans. If I invest £20 and keep back £80, I may appear to be inefficient in the economic establishment but I will survive as Mother Nature does. I have also used the example of kidneys. We have two of them but an economist would say that is inefficient and that we should only have one. A mathematical economist would go further and say that we do not even need one, that we could store one in the centre of London and go and use it when needed. But if unforeseen contingencies occur such as a bus or tube strike, then you would be in danger. It is better to have reserves.
Moreover, the environment, since the 1980s and the invention of the internet, has become much more complex to predict. Economic variables have in fact become completely unpredictable meaning that forecasts can be subject to enormous errors. Therefore, the less debt we have and the more resistant to forecasting errors that we are, the more robust we are in the face of Black Swans.
The second issue is size. The larger the company, the more Black Swan prone it will be whereas small and decentralized companies are resilient to the Black Swan. Again, we can learn from Mother Nature here. The largest animal on land is an elephant; if an elephant dies it does not bring down the entire ecosystem but if a bank dies, it brings down the entire banking system. I’ve used Mother Nature as a guideline on how to design society from a risk management standpoint. In an entity that is bigger, errors cost proportionately more and can be disproportionately more harmful than for smaller corporations. If you have a big company with a rogue trader, it will cost you a lot more per dollar than if you are much smaller because liquidating £50 billion costs a lot more per unit than liquidating a £5 million position. If you are an elephant you are going to take much more per liter than if you are a mouse.
There are advantages that are cosmetic in getting larger. Wall-Mart for example would appear to be more efficient than others but typically these sorts of firms do not survive. Look at the SP-500 in 1970 and again in 2000 and look at all these firms that got larger and achieved economies of scale and everything they teach you in business school that works, yet they did not survive. Economies of scale are misleading. The risk is disproportionate to economies of scale and wipes it out; the numbers show it.
Mergers can also make matters worse. When two companies combine expertise you might think they would grow and become a stronger unit and it is not necessarily due to managerial issues, it is often very simply the case that one random event costs them a lot more than when they were two small firms.
There is of course the issue of paranoia in regard to Black Swans and whether the fear can induce a form of paralysis which is more harmful to the event itself. While paranoia can indeed be an issue, we will never make it if this society can not become more robust to Black Swan events. The solution is therefore not about changing our nature and it is not about tell people to not take dangers seriously. The solution is to be protected as much as you can from these events.
Paranoia can pay in the long run. It is not necessarily irrational. But we have to get society prepared, beforehand, for mass irrationality. People argue against rumors, bubbles and emotion in the market. But the market is emotional. We need to build a system that can handle this; handle emotion, handle rumors and handle irrationality.
Rare, Black Swan events are in-computable. So my theory revolves around organic prevention; being prepared beforehand for events that are not seen. Maybe the best form of organic prevention is to be nice to your neighbors rather than having a big army. In the long run, everything that is Black Swan fragile will break. What we need is a commerce driven world rather than a top-down militaristic system.
The smaller the probability, the more you think it is predictable. This is why despite there being over a million economists on this planet, when I was shouting aloud that the system has taken risks, out of a million, only four or five were sympathetic. How many predicted the crisis? A handful.
Mother Nature has a lot of interconnectedness yet it survives because no land animal is bigger than an elephant. You cannot stop interconnectedness, it is too late, and we are not going to ban the internet. So we have to learn how to live with it instead. Interconnectedness has changed everything; what was once prone to forecasting is no longer so. We live in the new world, since the internet, and you have to accept the consequences and you cannot change the world, you have to become more resistant; through low debt and reduced size.
Social-economic life does not have the same properties as the engineering world where you can figure out margins accurately and effectively. Engineering has errors for ‘mediocristan’; I can measure the stable, I can measure the risks of a building collapsing, I can measure material, but I cannot measure the risk of future events in social-economic life.
To prevent Black Swans from society, let me summarize two simple rules; one, do not give the manager of a nuclear plant an incentive bonus based on cost cutting; two, the captain goes down with his ship. The first rule is about the avoidance of some risks and the second rule concerns the elimination of that form of capitalism that we have. From that we can derive a whole host of heuristics.
If you let the system work itself, the giants will self destroy. The regulations need to be done the right way. The regulators in Basle are the ones who help bankers pile up risk. We need regulations that help the small guy and help the economy to have protection against the risk takers.
In the ‘right’ society, banks would play a much smaller role. Lending and debt would be low. It is a big bogus argument that you need debt for growth. In 2000, we had technology companies experiencing a big equity bubble and Warren Buffett was complaining that these companies can raise £1billion but they cannot borrow £5million from the bank which he thought was a bad idea. First of all, it was not a bad idea and second, the minute companies mature and want to play the moral hazard game, then they start borrowing and bankers have not been there to help people who need the money (venture capitalists), bankers have been there to earn themselves, and other people, bonuses.
Look also at the Ten Principles for a Black Swan-Robust Society
Go here to look at our new eBook were you can read more about searching for negative and positive aspects with ideas and ways to deal with Black Swans.
Photo “Black Swan Head” by Tina Phillips
2 Replies to “Black Swan Robustness”