Pareto Optimality — Definition, Formula & Examples

Pareto optimality is a property of an outcome: it is Pareto optimal when you cannot make anyone better off without making at least one other person worse off. It is also called Pareto efficiency.

Let $X$ be a set of feasible outcomes and let $u_1, \dots, u_n$ be the players' utility (payoff) functions. An outcome $y$ Pareto dominates an outcome $x$ if $u_i(y) \ge u_i(x)$ for every player $i$ and $u_j(y) > u_j(x)$ for at least one player $j$ . An outcome $x \in X$ is Pareto optimal if no outcome in $X$ Pareto dominates it. The set of all Pareto optimal outcomes is called the Pareto frontier.

Worked Example

Problem: Alice and Bob must choose one of four outcomes. Each outcome is written as (Alice's payoff, Bob's payoff): A = (4, 4), B = (6, 2), C = (3, 3), D = (2, 6). Which outcomes are Pareto optimal?

Step 1: Compare C with A. Both players get more under A, so A Pareto dominates C. Switching from C to A helps everyone and hurts no one, so C is not Pareto optimal.

u_{\text{Alice}}(A) = 4 > 3, \qquad u_{\text{Bob}}(A) = 4 > 3

Step 2: Check A. To dominate A, an outcome must give both players at least 4. B gives Bob only 2, D gives Alice only 2, and C gives each player only 3. No outcome dominates A, so A is Pareto optimal.

2 < 4, \qquad 2 < 4, \qquad 3 < 4

Step 3: Check B and D. To dominate B, an outcome must give Alice at least 6, and only B does that. To dominate D, an outcome must give Bob at least 6, and only D does that. So neither B nor D is dominated.

\max(4, 6, 3, 2) = 6

Answer: A, B, and D are Pareto optimal. C is not, because A gives both players more.

Why It Matters

In the Prisoner's Dilemma, mutual defection is the predicted result, yet mutual cooperation Pareto dominates it. That gap is the key idea behind the game. In economics, the First Welfare Theorem states that competitive markets, under certain conditions, produce Pareto efficient allocations. Engineers use Pareto frontiers when they trade off goals such as cost against weight in a design.

Common Mistakes

Mistake: Assuming a Pareto optimal outcome must be fair or equal.

Correction: Pareto optimality only rules out waste. It says nothing about fairness. In the example, B = (6, 2) is Pareto optimal even though it is unequal. Giving one person everything is also Pareto optimal, because any change would make that person worse off. To judge fairness, you need separate criteria, such as envy-freeness or equal shares.

Related Terms

Game Theory — Field where payoff outcomes are compared for efficiency
Prisoner's Dilemma — Classic game whose predicted outcome is not Pareto optimal
Optimization — Pareto frontiers solve multi-objective optimization problems
Fair Game — Fairness concept distinct from efficiency
Borda Count — Voting method that satisfies the Pareto criterion