"On average across worker pairs, spectators implemented a Gini coefficient of 0.56 and redistributed $1.30 of the winner’s earnings to the loser. "

"In the efficiency cost treatment, the winner was determined by a coin flip and redistribution incurs an “adjustment cost” that reduces the total earnings available to participants. For every $1.00 reduction in the winner’s earnings, the loser’s earnings increase by only $0.50"

So out of $6, the average MBA of the 271 sampled redistributed $1.30 to the loser who was not obligated to receive any of it. So 22% redistribution. Less if the redistribution has costs. Seems fair?

I imagine that how the students behaved and what "seems fair" is heavily biased by the cultural context.

I agree that 22% redistribution seems about fair, but I'm biased by a lifetime of rates of taxation.

Something I'm very curious about is what should we be trying to optimize for (economic growth, sense of well-being, energy output, raising the economic floor, etc.?) and then what gini coefficient optimally produces that outcome.

I understand it's outside the scope of this study, but my bias is my curiosity for what should we be aiming at and what principles will optimally produce the result, so then at least we know in theory what we should be trying to do.