A fascinating new paper on ergodicity.
It's quite technical, so I took the freedom to explain here some of its findings and add on some of its applications.
twitter.com/ole_b_peters/status/1528646949947285505?s=20&t=jgUMYPOPXHhNEb6UoMxAqA
The paper answers the question of "why do people cooperate by sharing resources"?
Common answers include reciprocity ("I help you today so you help me tomorrow") or specialization ("I share my food and you share your tools").
Instead, the paper shows how cooperation is an advantageous choice even without considering reciprocity, specialization, and similar indirect benefits.
In a volatile world, cooperation increases growth simply by reducing the irreversible losses caused by volatility.
The paper offers strong mathematical demonstrations, but consider the following simple example.
If you and I are hunters, and do not share food, a series of unfortunate "no catches" might mean I suffer famine and die.
This applies even if hunting is an activity that, on average, sustains hunters.
Conversely, if we share part of our catches, we ensure that no one suffers famine unless we both have a long series of "no catches" – a quite unlikely event.
This is the advantage of cooperation: it reduces the likelihood of losses that absorb future gains.
I explain the problem of irreversible losses here, with a story about my cousin and his short career as a professional skier
twitter.com/DellAnnaLuca/status/1339621378765537282
There is the common belief that sharing resources (eg, taxes) is good both for the individuals who produce less than average and for societies as a whole but not for the top producers, who give more than they get back.
This paper shows how this belief is (partially) false.
If production rates are constant, then yes, sharing is bad for top producers.
And if tax rates are 100% (imagine communism), sharing is also bad for top producers.
But if tax rates are appropriate, sharing is good even for top producers – for it smoothens irreversible losses.
A simple way to think about it is: stock markets usually have positive long-term returns. This means that if you stay invested, you'll reap good returns.
But if a market downturn causes you to disinvest, you won't be able to benefit from the following rise in stock prices.
Irreversible losses absorb future gains.
Sharing resources mitigates irreversible losses, thus enabling people to benefit from activities that offer "high average growth rates over long-terms"
Two sentences from the paper particularly struck me.
"Good risk management does not merely reduce the size of […] down-swings. It also improves long-time performance"
For example, if an employee quits your company, the cost is not just the time, effort, and sign-on fees to replace them.
It's also the fact that whatever skills & knowledge the employee accrued, and could have produced returns, are now gone.
Also: "individuals poll and share resources because, over a temporal sequence of interactions, it is individually advantageous for them to do so […] group selection is not needed [to explain sharing] because the interests of the group and individual are aligned."
I encourage you to read the full paper, or at least the non-mathematical parts.
I will include some comments on it as a bonus chapter of my book "Ergodicity" (link in the tweet pinned in my profile).
Summary: sharing resources mitigates irreversible losses, thus enabling people to benefit from activities that offer "high average growth rates over long-terms"
In the presence of volatility, this is beneficial even for top performers.