Things that Gain From Disorder
Anti-Fragile
By Nassim Nicholas Taleb
“Wind extinguishes a candle and energizes fire. Likewise with randomness, uncertainty, chaos: you want to use them, not hide from them. You want to be the fire and wish for the wind…We want to survive uncertainty, and in addition have the last word. The mission is how to domesticate, even dominate, even conquer, the unseen, the opaque, and the inexplicable”
How?
What is Antifragile?
Some things benefit from shocks, randomness and disorder. However, we do not have a word for these systems — let’s call it Antifragile. Antifragile is different from robust. A robust object gets shocked and stays the same, the Antifragile gets better!
The Antifragile loves randomness,uncertainty and errors. Taleb writes that we are better at doing than thinking, thanks to Antifragility.
“I’d rather be dumb and antifragile than extremely smart and fragile, any time.”
“Instead of a discussion of risk (which is both predictive and sissy) I advocate the notion of fragility, which is not predictive-and, unlike risk, has an interesting word that can describe its functional opposite, the non-sissy concept of antifragility”.
“We can almost always detect antifragility using a simple test of asymmetry: anything that has more upside than downside from random events (or certain shocks) is anti-fragile; the reverse is fragile.”
Taleb classifies all systems as either Fragile, Robust or Antifragile — the triad. “The fragile wants tranquility, the antifragile grows from disorder, and the robust doesn’t care too much.” Fragile systems have unfavorable symmetry with more to lose than to gain.
The Triad
Asymmetry
Fragility implies more to lose than to gain, equals more downside than upside, equals unfavorable asymmetry.
Antifragility implies more to gain than to lose, equals more upside than downside, equals favorable asymmetry.
The Human Body
The human body, like almost everything in nature, is Antifragile. When you apply stress to the system it gets stronger. Of course, this is up to a point — but this doesn’t rule it out from being Antifragile.
“for instance, your bones will get denser when episodic stress is applied to them, a mechanism formalized under the name Wolff’s Law after an 1892 article by a German surgeon”
Prediction
By understanding Fragility we can build a systematic guide to decision making under uncertainty. It is easier to understand if something is fragile vs to predict the occurrence of an event which may harm it. Taleb argues that Fragility can be measured while risk (in the real world) cannot.
Antifragility provides a solution to the Black swan problem — “the impossibility of calculating the risks of rare events and predicting their occurrence”.
Instead of trying to predict events we should focus on making things more antifragile and to have things which do not fall apart when a mistake occurs.
“After the occurrence of an event, we need to switch the blame from the inability to see an event coming (say a tsunami or financial crisis) to the failure to understanding fragility, namely “why did we build something so fragile to these types of events; not seeing a tsunami or an economic event coming is excusable; building something fragile to them is not.”
It turns out that Netflix had made themselves “antifragile” by employing a tool they called “Chaos Monkey.” What Chaos Monkey would do was to simply regularly and randomly “crash” various Netflix servers. (“Crash” is in quotes because when it is being done on purpose by the machine owner, it is not clear whether it really should be called a crash or not.)
I just read about how great Amazon US-EAST crash of April 21, 2011 brought down most of their customers who depended on that zone, including big one’s like Reddit and Quora. But Netflix remained up. How did that happen?
It turns out that Netflix had made themselves “antifragile” by employing a tool they called “Chaos Monkey.” What Chaos Monkey would do was to simply regularly and randomly “crash” various Netflix servers. (“Crash” is in quotes because when it is being done on purpose by the machine owner, it is not clear whether it really should be called a crash or not.)
By continually crashing their own servers, the Netflix engineers could keep on learning how to keep uncrashed portions of their network up and running in the face of part of the network going down. And so when Amazon US-EAST crashed, Netflix ran on, unfazed.
Society Today
Taleb argues that society today has become increasingly fragile due to top-down systems being imposed and deprives of stressors. The greatest “fragilizer” of society is the absence of “skin in the game”. In which people get the upside of volatility (gains) but position themselves such that they do not get the downside — the Bob Rubin trade.
The Black Swan Fix
“Black Swans are large-scale unpredictable and irregular events of massive consequence…I have made the claim that most of history comes from Black Swan events while we worry about fine-tuning our understanding of the ordinary, and hence develop models, theories or representations that cannot possible track them or measure the possibility of these shocks.”
The central point of Black Swans is that “the odds of rare events are simply not computable.” The less common the event the less we know about it and its occurrences.
Instead of discussing risk which Taleb argues is a useless conversation he advocates for the notion of Antifragility as this can be measured. You cannot say that one remote event is more likely than another but you can accurately say that one system is more fragile than another.
The Antifragile: An Introduction
You are sending a package full of champagne glasses to a cousin in Central Siberia. You put a big label with fragile on the front. So what is the opposite of this scenario? Most people would answer robust/solid/resilient — however this is not truly the opposite. The opposite of “please handle carefully” is “please mishandle / please do not be careful”.
Anti-Fragile is best when harmed, robust is neutral as it remains the same and fragile gets worst when harmed.
Some Greek Myths
Damocles: The King Dionysus II has the fawning courtier Damocles enjoy the luxury of a banquet but with a sword hanging over his head tied to the ceiling with a single hair from a horses tail which will eventually break. Damocles is fragile. This breakage is a black swan. To continue success, what matters is the strength of the string.
Phoenix: The Phoenix always gets reborn from its ashes. It is Robust as it survives but doesn’t improve with such an event.
Hydra: Hydra is a serpent like creature that has multiple heads. When one is cut off, two grow back in its place. Hydra is anti-fragile.
Antifragile Responses as Redundancy
Nature likes to over insure itself as it understands that unusual/black swan events do occur. This is why we have two kidneys, extra lung capacity etc. Redundancy can be opportunistic as such extra strength can be used to some benefit later on — if you have an excess of electricity stored and there is a shortage you can sell at a large premium.
This is different to risk management professions who commit what Taleb calls the Lucretius problem in which someone will take the worst-case history event as the benchmark without realizes that this so-called worst-case event, when it happened, exceeded the worst case at the time. A turkey is fed for a thousand days by a butcher; every day confirms to its staff of analysts that butchers love turkeys “with increased statistical confidence”. Then the butcher will kill the turkey at the point that the turkeys confidence is at a maximum. Do not mistake absence of evidence for evidence of absence.
AntiFragility By Layers / Stress is information
The Antifragility of a system can come at the expensive of the fragility of its components. For example, the fragility of every startup is necessary for the economy to be antifragile. Errors convey information to the general group. If the Titanic had not sunk then we would have continued to build larger and larger ships with the inevitable next sinking causing even larger issues. Good systems allow small, independent errors to occur to allow the flow of information.
We should build systems, such as aviation, in which allow many small errors independent from each other, with these errors being negatively correlated — one error leads to less errors later as information is shared.
“If every plane crash makes the next one less likely, every bank crash makes the next one more likely.” with Banking systems being prone to contagion.
“My characterization of a loser is someone who, after making a mistake, doesn't introspect, doesn’t exploit it, feels embarrassed and defensive rather than enriched with a new piece of information”.
For an economy to be antifragile and undergo evolution, its units must be fragile and exposed to breaking. Organisms must die in order to achieve improvement or to avoid reproduction when they are not as fit as another. If we are “naive interventionists” we can starve a system of these stressors/information by intervening to heavily from the top-down (government bailouts etc.).
The economic system is inherently fragile. Unlike, say, a bunch of restaurants in an area (fragile individually, antifragile as a whole — unpopular or bad restaurants will go down in favor of better liked ones), globalized economic systems operate as one entity — errors spread and compound. It is not possible to have small mistakes and learn from them, in a manner like the global aviation community learns from one major plane crash.
We want people to take risks as long as people do not take the same risks and that these risks are small and localized while we do not want top-down structures which try and prevent volatility.
Iatrogenics — Iatrogenics is when a treatment causes more harm than benefit. This occurs frequently in medicine by doctors (naive interventionists) giving incorrect meditations with little understanding of second and third order side effects.
Crimes Against Children
Taleb writes that by not positively accepting stressors we are committing “crimes against life” for the sake of eliminating volatility.
By acting like soccer-moms and over-medicating/protecting children from variations in life we are harming them.
“If large pharmaceutical companies were able to eliminate the seasons, they would probably do so-for a profit of course.”
Evolution
Evolution benefits heavily from randomness and variation. The more noise, the more effect of the reproduction of the fittest and that of random mutations will play a role in defining the properties of the next generation.
If ten children are born into a stable environment, all ten will reproduce. If they are born into a volatile and difficult environment only the strongest will benefiting the species as a whole.
National Entrepreneur Day
“Someone who did not find something is providing others with knowledge, the best knowledge, that of absence (what does not work)-yet he gels little or no credit for it. He is a central part of the process with incentives going to the others and, what is worse, gets no respect.”
Taleb writes that societies should treat ruined entrepreneurs like we honor dead soldiers — there is no such thing as a failed soldier dead or alive.
Book II deals with how we are fragilizing social and economic systems by denying them stressors and randomness.
Crucially, if antifragility is the property of all those natural (and complex) systems that have survived, depriving these systems of volatility, randomness, and stressors will harm them. They will weaken, die, or blow up. We have been fragilizing the economy, our health, political life, education, almost everything … by suppressing randomness and volatility. … stressors. Much of our modern, structured, world has been harming us with top-down policies and contraptions (dubbed “Soviet-Harvard delusions” in the book) which do precisely this: an insult to the antifragility of systems. This is the tragedy of modernity: as with neurotically overprotective parents, those trying to help are often hurting us the most (see iatrogenics)
Extremistan vs Mediocristan
Bottom-up volatility not constrained by humans produces graph 1 which Taleb calls Mediocristan. There is constant volatility but over time it tends to average out. However, when humans cage systems we tend to produce graph 2 — long periods of time with low volatility and then black swan events which case huge swings.
The Civil Servant vs Taxi-Driver
Taleb describes two brothers. One is a Civil Servant with a flat, predictable income and another is a taxi-driver with a more volatile income. The Taxi-Driver complains that he has a riskier profession than his brother. However, this is not true.
Although the driver’s income changes he is far less susceptible to a black swan event which halts his income completely while through constant randomness and testing he improves through these stressors — continuously improving. The civil servant has one employer while the driver has many. If the civil servant punches his boss he will get fired, if a taxi-driver has a confrontation with a customer he will be able to survive until the next day.
The Signal and the Noise
Taleb describes how data can cause more harm than good.
“The more frequently you look at data, the more noise you are disproportionately likely to get hence the higher the noise-to-signal ratio….Say you look at information on a yearly basis, for stock prices, or the fertilizer sales of your father-in-law’s factory…the ratio of signal to noise is about one to one — this means that about half the changes are real improvements or degradation, the other half come from randomness. This ratio is what you get from yearly observations. But if you look at the very same data on a daily basis, the composition would change to 95% noise, 5 % signal. And if you observe data on an hourly basis, as people immersed in the news and market price variations do, the split becomes 99.5% noise to 0.5 signal.”
Benefits of Volatility
There a benefits to volatility with “variations acting as purges”. Small forest fires periodically cleanse the system of the most flammable material preventing accumulation. As a result, preventing small forest fires leads to one single much larger fire.
In the economy, a lack of variation can weaken businesses and can allow hidden risks to build up.
“Seeking stability by achieving stability has been a great sucker game for economic and foreign policy makers” — leading to Extrimistans.
Stoicism
Stoicism allows us to be antifragile by giving us a positive asymmetry with only upside and not much downside.
When Zeno of Kition suffered a shipwreck, he declared himself lucky to be unburdened so he could not do philosophy.
Success makes you fragile as you have more to loose than to gain. If gaining an additional amount of wealth would not benefit you but losing it would hurt you then you have an asymmetry making you fragile. To counter this, Seneca would write off possessions, so that when losses occurred he would not feel a sting.
Path Dependency
To start to mitigate fragility, one must understand path-dependency which states that what matters it the order of events themselves not just the events.
“The consideration of path dependence makes our approach simple: it is easy to identify the fragile and put it in the left column of the Triad, regardless of upside potential -s since the broken will tend to stay permanently broke….Notions such as speed and growth are empty and meaningless when presented without accounting for fragility. Consider that someone driving two hundred and fifth miles an hour in New York City is quite certain to never get anywhere….
If a gambler has a risk of terminal blowup, the “potential returns” of his strategy are totally inconsequential.”
Barbell Strategy
The Barbell strategy is based on playing it very safe in some areas (staying robust to negative black swans), and taking a lot of small risks in other areas (open to positive black swans), to take advantage of antifragility. While avoiding being “in the middle.”
If you put 90% of your net worth in cash or T bills, and you use the other 10% for extremely aggressive and risky investments, you can never lose more than 10% of your net worth, but you’re exposed to massive upside.
“This barbell technique remedies the problem that risks of rare events are incomputable and fragile to estimation error; here the financial barbell has a maximum known loss.”
The legendary investor Ray Dalio has a rule for someone making speculative bets: “Make sure that the probability of the unacceptable (i.e., the risk of ruin) is nil.” Such a rule gets one straight to the barbell.
Lecturing Birds on How to Fly
It took 6,000 years between the invention of the wheel and the wheeled suitcase. The first wheeled suitcase was invented in 1970 — after we put a man on the moon.
“This tells us about the way we map the future. We humans lack imagination to the point of not even knowing what tomorrow’s important things look like. We use randomness to spoon-fed us with discoveries-which is why anti-fragility is necessary….The story of the wheel itself is even more humbling than that of the suitcase: we keep being reminded that the Mesoamericans did not invent the wheel. They did. They had wheels. But the wheels were on small toys for children. It was just like the story of the suitcase: the Mayans and the Zapotecs did not make the leap to the application. They use vast quantities of human labor, corn maize, and lactic acid to move gigantic slabs of stone in the flat spaces ideal for pushcarts and chariots where they built their pyramids.”
Taleb writes that there is something sneaky in the process of discovery with most discoveries owing a lot to randomness. Randomness plays a role in the two levels of innovation: the invention and the implementation. Taleb writes that the second step is often the hardest with many inventions being just half-invented with no implementation.
Domain Dependence
You are escorted into a room. On one corner there’s a table with three items on it: a box of board-pins, a matchbox, and a candle. Your task is to attach the candle to the wall, so the wax doesn’t drip onto the table.
A psychologist named Karl Duncker first designed this experiment in 1945.
About seventy-five percent of the participants who take part in this experiment try following solutions.
First, they try to pin the candle onto the wall. It doesn’t work. Then they try to light the candle and use the dripping wax to attach it to the wall, but that’s usually not strong enough to hold the candle. So that doesn’t work either.
What about you? How would you solve this? Take a moment and think about it.
Very few people see the solution at once. Some people find it after only a minute or two of thought. Others see it after stumbling through several unsuccessful attempts. Most fail to solve it without some outside help.
Here’s a hint — think outside the box. A cliche, isn’t it?
Since ages, thinking outside the box is an overused, rather over-abused, phrase by every creativity expert. However, in this candle problem, you literally have to think outside the box.
Which box? The one that holds the board pins.
By the way, this is not a test of your intelligence so don’t worry if you couldn’t solve it. The experiment is meant to reveal how the human mind works.
Here’s one possible solution –
Take out the pins from the box and pin the box to the wall. Light the candle. Soften the bottom of the candle with a match, so that the wax begins to drip into the box, and place the candle inside the box, on top of the soft pillow of wax. Secure.
Most people in this situation do not see that something obvious — a box of board pins — might be something less obvious: a box and pins. Scientists have a name for this — functional fixedness.
Our knowledge about the utility of the box gets frozen by its domain, i.e., as a container to hold objects. In other words, the box containing the pins gets us hooked into the domain of “box as a container” realm. We fail to think outside that realm.
This limitation of our ability to pass the knowledge from one domain to another is called domain dependence where “Domain” is an area or category of activity.
Intelligence
Antifragility is often better than intelligence and we can combine it with trial and error to find things out as long as we can keep surviving to the next day (path dependence). We just need the intelligence to know when something great has been discovered.
“We need to exercise the option.”
Dis-information vs confirmation
“We know a lot more what is wrong than what is right…negative knowledge is more robust to error than positive knowledge…If I spot a black swan, I can be quite certain that the statement “all swans are white” is wrong. But even if i have never seen a black swan, I can never hold such a statement to be true…dis-confirmation is more rigorous than confirmation.”
The Lindy Effect
The Lindy effect is a theory that the future life expectancy of some non-perishable things like a technology or an idea is proportional to their current age, so that every additional period of survival implies a longer remaining life expectancy. If a book has been in print for forty years, I can expect it to be in print for another forty years.
The Soviet-Harvard Illusion
Real knowledge comes from the process of Random Tinkering (antifragile) → Heuristics (technology) → Practice and Apprenticeship –> Random Tinkering (antifragile) → Heuristics (technology) → Practice and Apprenticeship
Soviet-Harvard illusion is that academic knowledge is superior and that we must understand the mechanism in order to understand the effectiveness or phenomenology.
Taleb writes that a bird will fly first and then scientists will work out how not the other way around.
Taleb further discusses this illusion in Finance. Academics will look at traders in practice and then write theories. It is then deemed that traders are putting theories into practice!
The Agency Problem
“The worst problem of modernity is the transfer of fragility and antifragility from one party to another with one getting the benefit and another getting the harm…It is of course an agency problem. And the agency problem, is of course, an asymmetry.”
Taleb writes that our society lacks heroism which is the opposite of the agency problem with someone taking a personal downside risk to help others.
This is summed up in the Robert Rubin trade. Robert Rubin earned $120 million from Citibank in bonuses over about a decade. Citibank collapsed (Turkey surprise) but he kept his money. Unfortunately, the Hammurabi code used by the ancients is not followed in these scenarios.
“ If a builder builds a house for someone, and does not construct it properly, and the house which he built falls in and kills its owner, then that builder shall be put to death.”
