Book 2: A First Look at Agency
Chapter 1: Why Each One Should Eat His Own Turtles: Equality in Uncertainty
You who caught the turtles better eat them. This ancient quote means that you should assume whatever consequences of your actions.
Beware of people that give you advice “for your own good” (especially when unsolicited) as it is often good for them too…except that they won’t pay the price if something goes wrong.
-> selling and giving advice should be separated, since you have SIG when you sell, but not when you give advice.
This leads to an ethical question: should you disclose all of your stakes when selling to a customer?
According to the author, yes.
The ethical is always more robust than the legal. Laws come and go. Ethics stay.
Which brings us to asymmetry.
To what extent can people in a transaction have differences in the amount of information they each hold?
There are different answers.
1. In our world, less and less, as we are moving towards transparency due to the desire to stop cheaters.
2. For Greco-Roman philosophers, one philosopher applied total transparency, while another one advocated for legal transparency -> you had to be transparent as much as the law asked you to.
3. Shariah (Muslim law) forbids gharar, which can be translated as the inequality of uncertainty -> complete transparency is mandatory. If not, it is considered theft.
4. For Jews, you can’t change the price of products according to the net worth or motivation to buy from the consumer.
However, as we have seen, things seldom scale. Do we really have to apply ethics to everyone?
No.
We think local terms, not universal terms.
According to Nietzsche: Sympathy for all would be tyranny for you, my good neighbor.
Eg:
- When Athenians treat all opinions equally, it doesn’t apply to slaves.
- Romans stripped the citizenship of Romans that had married barbarians.
- Jewish ethics distinguishes between thick and thin blood. We are all brothers, but some are a bit more brothers than others.
We exercise rules in a context, but there is a limitation for scaling -> at some point, the rules cease to apply.
This leads to the following question: can you be both ethical and universalist? In theory, yes. In practice, no.
To quote the author:
For whenever the “we” becomes too large a club, things become abstract, degrade and each one starts fighting for his own interest.
Humans have a hard time seizing the abstract. In a way, the abstract is too abstract for humans -> being tribal (knowing where you apply ethics and where you stop) is not a bad thing.
According to Yaneer Bar-Yam, “better fences make better neighbors”.
-> scaling matters.
You simply cannot put Christians, Shiites, and Sunnis together and ask them to unite “for the country”. It’s been tried, and it has failed.
The globalists of our time blame people for being sectarian. Instead, they should look at what could be done with a sectarian mindset (eg: split people, give them each some land, and let them trade with each other).
The impossibility of scaling appears blatant when we compare villages to big cities.
In the former, people know and help their neighbors, and attend each other’s funerals.
In the latter, dealing with other people is based on different rules (mainly, you don’t deal with other people).
-> ethics is inherently local.
Why?
Modernity says that there are two units: the individual, and the collective -> it is the idea that you, being you, have exposure only to yourself.
In practice, this isn’t true.
You have exposure (SIG) to your family, friends, neighbors, etc.
But that exposure cannot be universal aka you cannot have SIG for everybody.
This is expressed by Elinor Ostrom, in an idea she called the tragedy of the commons and for which she won the Nobel Prize.
The tragedy of the commons works as follows.
Farmers or fishermen, as collectives, want to avoid overfish or overgraze not to deplete resources.
But the fisherman or farmer, at the individual scale, has everything to gain from overfishing and grazing – as long as others don’t do it.
Ostrom answers the following question: When do individuals stop protecting the commons to serve themselves?
-> when the group becomes too big.
This explains how tribes operate. When part of a tribe, you operate in a larger group than you, but narrower than everyone.
This is why socialism doesn’t work -> everyone loots for themselves.
The same is true for the opposite of socialism – a purely privatized system. People can’t work in such a system either.
Consider the following quote.
I am, at the Fed level, libertarian; at the state level, Republican; at the local level, Democrat; and at the family and friends level, a socialist.
-> the left VS right debate is ridiculous. It all depends on the scale.
Switzerland and other Germanic countries work well because the risk is shared among everyone in the population -> everyone has SIG.
Eg:
- In ancient Greece, when merchandise had to be thrown overboard to lighten a ship in a storm, all merchants had to pay, not only those whose merchandise had been thrown at sea.
- Same principle applied to caravans of merchants.
-> they had SIG.
So, what happens once you remove SIG?
It gets much worse. Let’s take the example of financial journalists.
Journalists, to avoid market manipulation, must give their opinion on stocks they don’t want to profit from.
This leads them to have impunity when they give bad advice, which comes at an ultimate higher cost than market manipulation.
Exposure often comes with a conflict of interest, which isn’t a problem, since the alternative (no exposure) is far, far worse
Exposure > conflict of interest. It is congruous.
On Medicine
Medicine is like engineering: apprenticeship-based and grounded in experience. There is exposure in medicine (both the doctors and you have SIG), but not fully in the agency effect between customer and provider.
Here’s why.
To “protect the patient”, the system tried to put more of the doctor’s skin in the game by establishing metrics to judge doctors by.
This shifted risks from doctors to patients because this is the wrong game – these metrics weren’t the right ones.
Eg: You get cancer and have a choice between curing cancer with radiation so you feel better during the next five years (but then worse) VS curing cancer with surgery where you may die the first five years but then live better.
Since these treatments are judged on a five-year scale, the doctor will choose the radiation and shift by doing so, the risk from him to you.
In a way, the objective of the doctor isn’t to cure you. It is to avoid a lawsuit if you die within the first five years.
This is why doctors give you medicine “preventively” (so that they’re not sued).
In the end, the medicine might actually end up killing you but the doctor won’t be condemned since he actually did something to save you.
-> both patients and doctors have exposure, and it is administrators that seem to be the cause for the trouble.
Like often.
Book 3: That Greatest Asymmetry
Chapter 2: The Most Intolerant Wins: The Dominance of the Stubborn Minority
What characterizes complex systems is that the system behaves in ways not predicted by its components.
Interaction > nature of the unit.
You can’t study an ant by itself, you must de facto study it in the colony, where the parts and the whole differ because of the interactions between the parts (called emergent property of the whole).
The rule we will discuss here is the minority rule, the biggest of all asymmetries.
Here’s the main idea.
Minorities (3% suffice) with enough SIG can submit the entire population to them, which is often not obvious.
All it takes for society to be influenced is a very small minority, intolerant, virtuous, and courageous.
Eg:
- In the US, all drinks are kosher to suit both the people that eat kosher, and the rest that doesn’t care.
A kosher will never eat non-kosher, but a non-kosher can (and will) eat kosher.
Let’s call the minority an intransigent group, and the majority, a flexible one.
Their relationship rests on asymmetry in choices.
On top of that:
- The geography matters, as there is a big difference whether the intransigent are in their own district, or mixed with everyone else. If they live in a ghetto, then the minority rule does not apply, because they don’t appear as a minority. But if they are mixed with the rest of the population, then it does.
- The cost matters. Eg: it doesn’t cost much to make all lemonades kosher, but it costs to make all buildings suitable for the handicapped. In the first case, the minority rule applies. In the second, it doesn’t.
Here’s how the dictatorship of minorities appears.
In the UK, 3-4% of the population are practicing Muslims. Yet, 70% of the lamb is halal.
Let’s visualize this principle.
The method we used to analyze this is called renormalization group.
Four boxes, grouped together, contain each four boxes.
Assume the smaller box is a family of four people. One of them, say, the 16-year-old daughter, is the intransigent minority and only eats non-GMO food.
That person manages to impose non-GMO food on the whole family (the tiny square is now red).
Now, the family goes to a barbecue -> everyone eats non-GMO to adapt to them. Then the local grocery store, learning that some families have become non-GMO, also becomes non-GMO, etc.
For most things, the intolerant will eventually triumph over the tolerant, since they refuse to adapt, while the latter accepts to adapt.
This is what explains the triumph of Islam.
In Islam, if a non-Muslim marries a Muslim, the non-Muslim must convert. Once you’re converted, you cannot go back to not being Muslim.
If a child is born off a Muslim and non-Muslim, it is de facto Muslim.
Islam’s rules were made so that it spreads.
The Jewish rules are the opposite. One can be Jewish only if the mother is Jewish.
Hence, Judaism doesn’t spread.
Other examples where intolerant minorities impose their rules concern:
- The banning of books
- The formation of moral values in society
- Civil rights
- Christianity in the Roman Empire. Christians refused to share Roman gods, while Romans shared everyone’s gods.
- Political entities: If a 3% minority in a country imposes its rule to the entire country, and if the entire country is around 3% of the EU, then it may impose its rules to the EU.
-> outcomes are more stable under the intransigent minority rule because the rule is often black and white and binary -> low variance in results.
Which leads to the question: should intolerant minorities be tolerated by tolerant people?
No.
If we let the minority rule do its work, it will eventually destroy the world.
So we need to be more than just intolerant with intolerant minorities, and cast them off entirely.
When it comes to the place that the West gave to some intolerant minorities like the Salafists, they are currently committing suicide.
Markets
We could say that markets aren’t the sum of market participants, but that the price reflects the activities of the most motivated buyer and seller.
The price can drop sometimes by 10% because of a single seller.
-> markets react in a way that is disproportional to the impetus.
Eg:
- If someone screams “fire” in a movie theater, everyone will run out.
Science operates on the same principle.
Science isn’t the sum of what scientists think, but it works like the market: once you disprove something, it is wrong.
Had science been built on a majority consensus, we wouldn’t have evolved.
Revolutions are similarly driven by a small number of stubborn people.
And the economy itself, rests on a small number of hyper-active people – entrepreneurs (Pareto’s law).
Conclusion: society doesn’t move forward by consensus. Only a few people are needed to disproportionately move the needle.
Asymmetric rule + soul in the game = great results.
And since asymmetry is present in pretty much everything…
Appendix to Book 3
The average behavior of the market participant will not allow us to understand the general behavior of the market.
-> markets are not the sums of average individuals.
By the same token, society is not the sum of average individuals, because individuals don’t act like individuals when in a group.
-> The psychology of the individual showing “biases” does not help us understand automatically the collective behavior of these individuals or the behavior of the group.
People don’t live alone, so restricting behavior to one person only cannot apply to reality.
Furthermore, groups are units of their own (things don’t scale, remember?).
A group of 10 ≠ a group of 200k.
The bigger the group, the more possible interactions, the more difficult it is to understand and predict based on information obtained about one single person by himself.
-> this is called the curse of dimensionality.
Side note: this is the same thing for the brain. Understanding how all parts work independently won’t help you understand how they work together.
Zero Intelligence Markets
The underlying structure of reality matters much more than the participants in the structure (it’s about the system, not its participants).
-> under the right market structure, a collection of idiots produces a well-functioning market.
In fact, it produces a similar market than if the participants were intelligent.
-> the invisible hand theory appears not in sync with reality.
Furthermore, the irrationality itself of some of the market participants may be needed to make a market efficient.
As a result, it’s not individuals that choose where they go, but the market.
-> once you provide a good structure, you can leave people alone and they will behave as they should.
Book 4: Wolves Among Dogs
Chapter 3: How to Legally Own Another Person
In the early days of the Catholic church, a group of begging monks who depended solely on the population to survive, were banned from the church (until they disappeared completely).
Why?
Because they were free.
Complete freedom is the last thing you want if you have a religion or a company to run.
-> any organization wants the people part of it to be deprived of a part of their freedom.
How can you own people?
- Psychological manipulation and conditioning
- Force them to have SIG so that they have something to lose if they disobey
Let’s imagine you run an airline with one plane. You hired your pilot as a contractor. One day, your pilot tells you he can’t fly the plane because an Arab sheik paid him more money for a party in Vegas.
Unfortunately, you can’t find anyone else to pilot the plane. You know you will go bust. Your company is over.
Suddenly you realize that your pilot didn’t do something that employees do.
Contractors are free, but employees aren’t. Had your pilot been an employee, he wouldn’t have behaved in this way.
Employees have exposure -> employees are usually risk-averse -> someone who has been employed for a while is giving strong evidence of submission.
Submission is shown when the employee:
- Works for someone else for 9h/day every day.
- Does not choose his own schedule.
- etc
-> an employee is an obedient housebroken dog.
Furthermore, the longer someone stays an employee, the more they are emotionally invested, the more they will be diligent.
The thing about employees is that they lower your tail risk because you own them. They also think you lower theirs…but do you?
Today, companies go bust faster than ever before -> people get laid off faster too.
In the past, company men would often do their entire careers in the same company. In the 1990s, as Silicon Valley was rising, many companies started laying off their people who could not find work anywhere else.
The company men are now gone. People are no longer owned by companies, but by something worse: the idea they need to be employable.
Meaning: it’s not that they cannot anger their company -> they cannot anger the entire industry, or they will never find work again.
An employee is someone you will never find in history books because they are designed to never leave their mark on any events.
Let’s have a look at the theory of Ronald Coase.
An employee is more valuable inside the company than out -> he is more valuable to the employer than to the marketplace.
A free market is a place where forces determine specialization and information travels via price points. Understand: people fight to get the best, which automatically creates velocity (people buying and selling) in the market.
But within a company, these forces are lifted because they cost more to run than what they bring. Understand: employees have no incentives to work since they get paid anyway.
Reminder: contractors have SIG doing good work as it influences their salary, but they can leave at any time if they find better opportunities.
Employees have SIG doing the bare minimum not to get fired, but you can force them to do stuff they don’t want to do as they could get fired otherwise.
-> hire the right number of employees and the right number of contractors.
The more skin someone has in a game, the more you can count on/control them.
Eg: The Romans had a slave to manage the family money.
Why? Because the slave could be killed if he made a mistake, while the free employee could not -> the slave had more exposure -> the slave would be extra careful not to make mistakes.
An employee is a weird form of slavery.
The best slaves are the ones you overpay, and they know it, and they are terrified of losing their salaries as a result.
This is why companies pay expats lots of money. Expats are far away -> expats are more free -> expats receive more money so they behave.
Freedom is not free
In the tale of the dog and the wolf, the dog brags about his house and unlimited food. The wolf is about to join him, until he notices the dog collar and asks what it is.
When he gets it, he runs away.
Freedom entails risks.
-> would you rather be a dog, or a wolf?
Whatever you do, don’t be a dog claiming to be a wolf.
Furthermore, the dog situation is false stability. If the owner leaves, the dog dies.
The wolf can always survive.
While we have established that employees are dogs, there is nonetheless, a category of employees that aren’t slaves. Those that don’t give a f*ck about their reputation.
Those are the traders and salespeople within companies, that would cause a higher loss to the company than they would suffer from themselves if they were to leave.
Loss Aversion
What matters isn’t what a person has or doesn’t have; it is what he or she is afraid of losing.
The more you have to lose, the more fragile you are.
The opposite of a slave is an autocrat.
Consider the difference between a Western head of state who has to give feedback to the press, political parties, and likely, other nations VS…Vladimir Putin.
One does what he wants and has a f*ck you money attitude, while the other, doesn’t.
-> it’s easier to trust the autocrat than the elected official, simply because the elected official isn’t free.
Let’s speak of another principle.
People whose survival depends on qualitative job assessments by someone of higher rank in an organization cannot be trusted for critical decisions.
The employee has no interest in the company, but has interest in keeping his job, and will make decisions in this direction.
The idea is that while we know when some courses of action are absurd, it is often easier to continue on this path than to stop.
The Vietnam or the Iraq wars fit this case.
In the latter, the people in charge did not want to take the risk to go to the source of terrorism (Saudi Arabia) and potentially miss on oil.
Same thing happened in 2009 with the banks, when no one went to prison, and when almost all of them were in fact, saved.
Chapter 4: The Skin of Others in Your Game
Imagine you work for a company and discover the company poisons thousands of people each year that die as a result.
If you blow the whistle, you may have to wait a long time before being believed, and your friends and family will suffer, and you will lose your job forever.
If you don’t, you lose sleep at night.
What do you do?
People with that volume of exposure (families, mortgages) are much easier to control, which is why corporations prefer people with families.
Most heroes (James Bond, etc) don’t have any families, so they can do “the honorable thing” at their expense only, and not sacrifice the 16-year-old vegan daughter.
-> to make ethical choices, you cannot have dilemmas between the particular (friends and family) and the general.
Intellectual and ethical freedom requires the absence of the skin of others in your game, which is why real free people are rare.
Most of the time, when you make powerful enemies that cannot pressure you because you are independent, they will pressure your close ones.
To be free of conflict, you need to have no friends.
Let’s analyze this one, taking as an example a suicide bomber.
How to Put Skin in a Suicider Bombers’ Game
The problem of suicide bombers is that they have no exposure since the end of their means is their own.
In the case of Hammurabi, if a house collapsed and killed the firstborn of the owner, the firstborn of the architect would be put to death too.
As we can see, the symmetry is specific, but the architect, while being solely responsible, won’t be alone to suffer.
In any way, this principle can be used in the case of terrorists. If we can convince them that their own death isn’t the end of their struggles, they might think twice before blowing themselves up.
Eg:
- impose financial penalties on terrorists’ families.
