SCIOPS 01.22: Destructive Equilibrium
I’ve been reading about game theory this week, and find myself curiously attracted to the concept of “Schelling points.”
Take a look at Seattle’s skyline:
If I tell you I’ll meet you in Seattle on 28 August, but gave you no other instruction, how would you find me? Of all the places in the city, when and where is the best probability of our meeting up?
If you said “the one like a hamburger on a stick”, then we’ll find each other. Especially if you picked noon as a meeting time – I’ll aim for noon, since it’s singular and precise. Sunset would also work, but it’s less specific. Maybe I’ll stand under the Space Needle from noon to sunset just in case. Of course, I’ll be surrounded by every other tourist in Seattle for that half-day, so if you do show up at sunset you better be ready to buy me a drink.
The Space Needle is a Schelling point , or focal point. It stands out from its surroundings, it’s unique and singular. Similarly, noon is a focal point in the day.
In game theory, Schelling points are important for finding what’s called a Nash equilibrium: an outcome in which neither player regrets the choice they made. If we both choose to go to the Needle at noon, we’ll find each other, and neither of us will regret having chosen to go there. If we each choose to give up on the meeting and hit a bookstore instead, we’ll each be glad we did once we know that the other wasn’t trying to meet up. These are Nash equilibria.
But notice that if I go to a bookstore, and you go to the Space Needle, you’re going to be waiting there all day for someone who’s not coming. By sunset, when you know I’ve stood you up, you’ll wish you had made a different choice. We are not in equilibrium.
However nice “equilibrium” might sound, it’s not always the ideal outcome. Remember, game theory is a product of the RAND corporation and the Cold War. Mutually Assured Destruction is a Nash equilibrium – if a nuclear power decides to bomb us into the ground, we’ll not regret having simultaneously chosen to vaporize them as well. John Nash proved that there’s at least one equilibrium in every game, but not only one: often there are better and worse choices by the metrics of all players.
In the wired world, a lot of new equilibria appear around the focal point of “internet”. For example, say I’m trying to get across Seattle for an important meeting at noon at the Space Needle, and you are unemployed and have a car payment coming up. I can pay you to use your car to drive me where I’m going. How do we find each other?
In the past, this was a tricky phenomenon. You either have to have a cab medallion and a phone number, or I have to see you driving by and hitch a ride in some convincing manner. But now there’s Uber. I can go to internet and search for a ride, and you can go to internet and find my offer. You give me the ride, I pay you, and everybody is happy. Especially Uber, because they get a percentage of the fare at almost zero cost to them. They’ve created a Schelling point, so they can charge rent to everyone who uses it.
Capitalism is a Nash equilibrium, where everyone’s incentives align toward getting the most revenue at the least cost. It’s obviously not perfect, though, as so many factors are left out of the equation. The profit motive has no concern for long-term damages to the ecosystem, for instance, because the player who that will hurt most aren’t alive yet to make moves in the game. This only gets worse with internet business, which take advantage of the worldwide flow of data to reduce costs and increase rents across the globe.
This latest mutant form of economics, the Uber/Amazon/Faceboot model, is called “ platform capitalism “.
The most profitable play is not just to find the focal point of internet and use it for meeting mutual needs; it’s to OWN the focal point and charge a cover fee for entry. The Space Needle Corporation exploits this in meatspace, but the global flow of internet provides much bigger opportunities for gatekeeping. In fact, the platform is by far the most profitable business model in 2017 – all the top earners use this monopolistic model.
What can we do about it? What other equilibria exist in the slippery new internet world? I have some ideas that I’ll share with you next week. In the meantime, you can send me your thoughts and links by responding to this email. Or we can always meet in Seattle – you know where to find me.
As always, thanks for reading.
is a cogsec newsletter. If you liked this issue, forward it to a friend. If you’re not signed up, you can sign up at
. Reflections and criticism can be levelled by responding to this email.