The Pareto optimum is defined by the fact that no other constellation than the Pareto optimum opens up a higher payoff for even one party.
In the prisoner’s dilemma example, the equilibrium with payoffs (2, 2) is stable, but not efficient. Therefore, it is not Pareto-optimal. In contrast, the constellation with payoffs (4, 4) is efficient and therefore Pareto-optimal. However, as we have seen, it is unfortunately unstable.
In decision situations in business practice, there may be a tension between an inefficient Nash equilibrium and an unstable Pareto optimum. In many situations, however, there are proven ways to enforce the Pareto optimum. Within firms and in value chains, for example, enforceable agreements can be made that allow relatively safe exploitation of Pareto-optimal equilibria. In competitive constellations, however, antitrust law speaks against binding agreements to jointly choose the Pareto optimum. Lack of trust leads to the emergence of an inefficient Nash equilibrium.
Problems of the common good are also more likely to result in an inefficient Nash equilibrium, because no one wants to contribute more to the common good than they expect others to do.