[azza2110]

I just ran it at 1% instead of $1.

On only 5% of days was the spread greater than 80 -> 120.

[lou1306]

I went a little further and ran a simulation [^1] with random payments in the range from $1 to a fixed percentage of the payer's wealth.

When this percentage is in the range 2-10%, the wealth inequality is significantly smaller than in the base case (i.e. all payments are $1). Then it starts growing again.

[^1]: https://gist.github.com/lou1306/1041ed6cd4eed433cfabf45f666b...