> if it was possible to determine ex ante what benefits society the most, it wouldn't even be a contest, of course we could all agree to do that, it's almost tautological.
The prisoner's dilemma is a simple example of a situation in which cooperation would be the ideal solution for everyone, nonetheless players tend to behave selfishly.
The problem with the current system isn't that people act selfishly i.e. in a way that creates most value for themselves.
The problem is that the selfish actions of some actors(the figurative 1%) count for overwhelmingly more than others. And a lot of actors don't have any say at all(those below the poverty line). In such a system, the selfish 'demands' of most other actors don't matter.
Say on HN, there were a few people (type A) that could instantly give 1,000,000 points to any comment or post and everybody else could give only 1(type B). A post could provide value for 100,000 type Bs and yet some other post that type A liked would always rise to the top. If this were true, HN would be a very boring place to be.
The goal should be to empower all actors to act selfishly - perhaps not equally, but atleast in a 'reasonably' proportionate fashion.
This holds for rare, luxury goods and limited markets (like housing). But the concentration of wealth in a minority AFAIK doesn't seem to bid up large markets like corn or paper.
Consider the example given by creamyhorror below - the influence of money in the political process.
What if besides having a disproportionately high number of points on HN, you had the power to influence the creators of HN so that a percentage of all the points on HN went to you? Or, what if you had millions of points on HN and were able to convince the makers of HN that you should be able to vote on HN with only 0.5 points per vote?
Yes, because the players do not have complete information. The correct strategy is only obvious to those outside the experiment.
When you extend the concept to society as a whole, we are all players. No one is outside it. That's why I agree with HSO that it's not possible to state absolutely what would best benefit society.
> The correct strategy is only obvious to those outside the experiment.
Wait, are you suggesting that cooperating in the prisoner's dilemma is the correct strategy then? Because in fact defecting is always a best response to any of the opponent's choices, even if you are aware of the rules of the experiment.
The prisoner's dilemma is a simple example of a situation in which cooperation would be the ideal solution for everyone, nonetheless players tend to behave selfishly.