Antifragile meets fragile reality
Nassim Taleb has written a book called “Antifragile — Things that gain from disorder”. Nassim claims in his book that there are things that have upside and very little downside. The name seems fitting, fragile things have huge downside, little stress and it is broken, lots of stress and it is annihilated. Fragile things are not sturdy; they are delicate and highly vulnerable. Antifragile is the opposite, a little stress, and it gains, lots of stress, abuse, harm, trauma, mishandling, accidents and it thrives.
In this article I will explore why literal antifragile is mostly a myth and why it seems my favourite author has been fooled by randomness.
Fragile and Antifragile
Nassim used champagne glasses as an example in his book. Champagne glass is useful or it is not, even if it is a little bit chipped it will most likely be thrown in the bin. Imagine you are moving your house and you have placed 5 champagne glasses in the back of the van. Unfortunately, or maybe fortunately, you are moving 400 miles away and roads are bumpy and you have packed these glasses carelessly as you never subconsciously liked them anyway. At the end of this trip, you might find that you only have 2 champagne glasses remaining. The only way for champagne glasses to trend is towards zero. It is guaranteed that over time they will cease to exist. I want to stress this, for true fragile things, there is only one way, that is down, towards zero.
It stands to reason that the opposite of fragile is something that is moving away from zero in positive direction and is hard to destroy. To use Nassim’s mythology example of Hydra the water monster, for every head chopped off, the Hydra would regrow two heads, however it can still be killed in a specific way. So the expected literal antifragile time series graph would look something like this:
An awesome way to imagine antifragile that is more contemporary than Hydra is nanobots from “The Day the Earth Stood Still”.
My above explanation of literal antifragility, is most likely not the type of antifragility that Nassim is thinking of, more on that later.
Next section for comparison will briefly explore adaptive, robust, resilient, complex and risky and opportunistic systems.
Adaptive not Antifragile
During 2021 I have trained for a marathon. Below graphic shows my progress, I am not a very fast runner. Training for a marathon is a lot of work, it puts lots of stress on your body. I did approximately 41 runs before the big day. During this I was able to increase my endurance from running 45 minutes to around 2 hours and 30 minutes. My speed has plateaued around 6.3 miles per hour average between short and long runs.
Sure, I felt lighter on my feet and when I was running shorter distances I felt fast, averaging 6.7 miles per hour, for me, that is quick. However there is nothing antifragile happening here. Just a human body, adapting to temporary stimuli and sadly, this adaptation is limited by my genetic makeup.
Some people might wonder, why is this not antifragile? Remember that antifragile is supposed to be the opposite of fragile. My gains plateaued at 6.3 miles per hour average, that was my physical limit. My improvements did not just go up, there was progress, but also there were regressions. There was also plenty of downside, time, potential injury and maybe in alternative reality I have died during the marathon or got hit by a car while training. These are characteristics of a bounded complex adaptive system.
Our body does need variance as it pushes us away from physical atrophy and degeneracy.
Robust - Slow deterioration
Certain objects such as machinery and even books can be classed as robust. Robust objects experience relatively long deterioration. Depending on the type of the object, the deterioration speed would look different, but it is there.
I know from personal experience that once a robust object such as a book is dropped into the sea it does not return to its original quality over time. It just keeps weathering away, slowly, until one day it reaches its final days and pages are no longer held by the book spine.
Resilient - Bounce back
Organic and artificial systems can be resilient. For something to be resilient it needs to be able to return to the previous shape after some traumatic event such as bending or stretching. So literary, it means to bounce back. In nature, when I get my tree pruned in my garden, nature recoils with branches, and once again I need to call the tree surgeon the next year to get it pruned.
In artificial systems we add redundancy so that we can fail over failed parts and come back to the original state of operation. Resilience does not increase capacity, it just ensures that everything comes back to the original capacity.
Positive and negative feedback loops
In complex systems there is a fascinating phenomenon called positive, negative feedback loops and emergence. These loops are multiplicative and they are caused by physically diverse agents interacting with one another with limited information. This can create frenzies that can be and appear random.
Let’s take a recent example, crypto, you get some crypto and tell your friends about it and how much money you are making. Your friends get into crypto and they tell their friends about their profits. Before you know it there are lots of people trading crypto. This right there is a positive feedback loop. Unfortunately if something is exposed to positive feedback loops it is also most likely exposed to negative feedback loop see below.
In Nassim’s book “Fooled by Randomness” he explains how simply being lucky once, can create more luck in the future as more doors open up as you leverage path dependence. Theoretically speaking a bit of luck can create order of magnitude more luck, creating your own positive feedback loop.
Risks and opportunities
It should go without saying that the best option when it comes to risk and opportunities is to pick “low risk and high opportunity” option. Terrible alternative would be “high risk and low opportunity”. In other words the best option should have high pay off and low penalty if something goes wrong and not low pay off and high penalty. Penalty should be minimised/curbed and opportunity maximised.
In the previous section I have mentioned that luck can create lots more luck. Well, opportunity can create more opportunity. Same can be said for risk, certain risks when materialised create more risk, also they can compound, that is some of them can happen at the same time and feed off each other. Nassim warns us of hidden risks and hidden opportunities, see below.
All pain, very little gain
What antifragile is talking about really is exposing yourself to positive feedback loops or compounding opportunities with very limited risks.
If you are an author your book might start trending among the population, creating a sales feedback loop . This will inevitably give you new opportunities and exposure further fuelling other opportunities. This could be akin to Figure 9 right. You have toiled day and night, mostly night, as you have a day job, for a long time and then, jackpot.
If you do not want to surf that positive feedback loop or opportunity. You can lead a successful life by protecting yourself by minimising your exposure to compounding risks, and negative feedback loops. For example by not having too many debts, having savings, needing little to survive and never investing more than you can lose. This could be akin to Figure 8 left.
I am sceptical of the antifragile sentiment in Nassim’s work. Reason is that we just don’t know what will experience the positive or negative feedback loop, what will trend tomorrow or 2 years from now. Sure, fragile things are guaranteed to blow up, maybe not this year, or next, maybe not even in our lifetime, but they will, that much is true, they must move towards zero. A lot of analysts knew that Tesla stock is fragilish, it has experienced phenomenal positive feedback loop, it was/is drastically over priced. Many investors have bet against it and lost huge. This is the problem with betting against fragility in a complex environment and assuming you are antifragile as result. There is no free lunch, you are still exposed to risk there is still downside, you are still fragilish and luck still plays a role.
History is littered with writers who were poor during their lifetime but became famous after they have died. I am sure that there are many amazing writers who have died and were never discovered even after their death. Same for scientists and researchers who made major contributions but were not appreciated and rewarded. Contributing something to humanity does not mean you will get to ride some version of positive feedback loop and get presented with opportunities. Figure 9 right, is called hidden high opportunity because for most entrepreneurs, writers, researchers they just don’t know if they will ever hit that hidden opportunity/reward. However if we do not try it is a guarantee you will not even be exposed to the possibility of it.
Antifragile?
It is time for us to come back to the beginning and re-ask the question. Is antifragile a myth?
Organic systems are mostly adaptive, can be fragile and sometimes they are resilient. Remember my marathon training? Physical objects are either fragile, robust or resilient. One thing seems to match the pure antifragile definition is information and subsequently, knowledge.
Information is an arrangement of representation. Information has huge upside, it loves stress, abuse, harm, trauma, mishandling and accidents. Any form of interaction with other information or the world gives birth to new information. Newly obtained information can entwine with other information creating whole new information in the form of discovery, insight and perspective. This process is multiplicative and it feeds on itself.
Information needs a host to exist, it is sheltered in our brains. It does not need to make any effort to multiply, it physically has no downside. It is not exposed to the harshness of reality. We need information to function in the web of our highly complex culture, and it needs us to feed our collective cultural consciousness by using us as data miners of our physical reality. At times information controls us more than we control it.
We should invest in quality information, learning, how we learn and create novel information.
Nassim vs Nassim
I have read most of Nassim’s Incerto series, I really enjoy his work. I have reread his “Fooled By Randomness” more times than I can remember. In my opinion it is a fantastic introduction to complexity, risk and variance in the markets. Nassim said in his randomness book, that in certain writing you can tell more about the writer than about the actual subject they are talking about. For me, this is the case with this book. Let me give you a few examples and add my own bias into the mix.
Taxi Drivers
Nassim has destain towards corporate employment, contempt is felt through most of his work. He thinks that people on corporate payroll, who get a monthly wage, are fragile and taxi drivers are resilient. A lot of corporate jobs are safer, as employees are protected by employment rights. This is not the case for the majority of self-employed individuals, including taxi drivers. A lot of self-employed individuals struggled during COVID-19, it is likely that corporate employees have not struggled as much.
Nassim’s issue with corporate staff is their lack of skin in the game. Most corporate employees do not face that many risks in their employment, while at the same time they have access to a lot of opportunities. Also if they do get fired, they can find another job relatively quickly, not impacting their livelihood for too long. Corporate employees, who are willing to adapt and have a slight workaholic slant can also find lots of opportunities to increase their monthly earnings. This slightly workaholic person can experience “low risk, great opportunity with hidden high opportunity”. Not sure if the same upside exists in the taxi world. This is the UK perspective.
I think it is safe to say that both taxi drivers and corporate employees can adapt and, dare I say it, corporate employees are safer and can be better off when it comes to earning potential and job safety.
Writers and artists
If I beat up the economist publically, what would happen to me? …Luca thought for a second, well, it is not like he would like me to do it, but, you know, it would not hurt book sales. Nothing I can do as an author that makes it to the front page of Corriere della Sera would be detrimental for my book. Almost no scandal would hurt an artist or a writer. — Antifragile
Most books don’t get read, also the amount of books you need to sell to actually make a living is depressing. Which means a lot of writers and I am assuming the same for artists, need to have day jobs to have some sort of reasonable income while they are developing their art.
Using notoriety as a tactic can close doors with publishers, speaking engagements, block you from certain institutions, impact your day job and so on. I understand that Nassim is looking to create a positive feedback loop, maybe even a negative one with this tactic, Figure 7. This might be a reasonable approach if you are rich or famous, so you have little risk exposure and or lots of opportunity to gain from. However this is not true for the majority of the population.
Most writers if hit by a negative feedback loop or by a compound risk from notorious behaviour might struggle to recover from it. So creatives are exposed to a lot of risk just like the rest of us.
Conclusion
Antifragile seems to say that you should pick options in life that are “low on risks and high on opportunities” and stay away from “high risks and low opportunities”. Also, more importantly, leverage asymmetry. Try to find opportunities where opportunity has a possibility of a positive feedback loop or compounding.
This is great advice, however is this new? It all feels overly optimistic. Can you actually become antifragile? Of course not, but you can try to become less fragile, but that is not the same thing. Also name, antifragile holds too much mystique, because it is not the logical opposite of fragile.
We just don’t know where there will be a positive/negative feedback loop or if opportunities will compound. If we did, well, we would all take advantage of them and it would not be such a special thing. A lot of it is luck and we give it backwards explanations because we want to feel smart and in control and not just fooled by randomness.
Personally, I will look out for compounding opportunities and try to reduce my known compounding risk exposure as much as I can. I will not chase the positive feedback loop unicorn. I will try to learn as much as possible and consider Charles Bukowski’s wisdom of “Don’t Try”.
