Pareto Efficiency
Your housemate came home...to you taking your first bite into a chocolate bar. You make eye contact. You know he loves chocolate. Awkward. You could give him some of your chocolate bar, but then you would get less chocolate than you had planned for. Dilemmas, dilemmas.
Good news: you don’t have to give up any of your chocolate. Just tell your housemate you’re being Pareto efficient, or Pareto optimal.
Pareto efficiency is when resources that are distributed to players can’t be moved without hurting at least one player. If you gave him some of your chocolate, sure, he’d be better off, but you’d be worse off. You’d only give him some if it didn’t detract from your payoffs.
Bad news: Pareto efficiency has nothing to do with fairness and everything to do with efficiency. Remember the production possibilities curve (PPC)? That’s where you have two goods you could make with the same inputs...like cookies and brownies. The curve shows all the options (cookie, brownie, etc.) where you’re fully utilizing all of your inputs. That makes all points on the PPC Pareto efficient.
The ideal Pareto efficient outcome is when everybody is better off working together. When the choice is obvious for both sides (like...both sides gain brownies), then it’s the Pareto efficient choice.
The trouble comes when players are incentivized to not cooperate. If there was one, lone brownie between you and your roommate, you both could share and get half each, one of you could grab it with the other getting nothing, or both of you grab it, and both of you get nothing since it ended up on the floor. This is why Pareto optimality is not Nash equilibrium.
In this brownie game, you might find that both of you go for it and grab it. Both of you are thinking “I’ll get the highest payoffs if I grab it first!” Too bad, since both of you grabbing it means neither of you get anything. Both of you grabbing would be a Nash equilibrium, where it doesn’t make sense for either of you to do any different. Both of you cooperating would bring you to a better place: a Pareto optimal place.
This is why Pareto optimal is sometimes associated with a “social optimum,” since it’s socially optimal in the sense that you can’t move any resources around without at least one player being worse off. If the resources weren’t distributed evenly to begin with, then this “social optimum” might not look fair, but it is Pareto efficient.
Life’s not fair, and neither is Pareto efficiency.