In high school you often face decisions such as how to divide study time across subjects or which tactics a sports team should choose while considering opponents’ reactions. Economics calls this “strategic interaction” and analyzes it with mathematical game theory. A Nash equilibrium is defined as a situation in which no player can gain by changing actions unilaterally, and it applies to topics from price wars to nuclear deterrence. Real negotiations, however, often take place in sequence and under uncertainty about the opponent’s information. Harsanyi modeled such incomplete-information games with Bayesian probabilities and introduced the “Bayesian equilibrium” where players have types and beliefs. Selten proposed “subgame perfect equilibrium,” a refinement that eliminates unrealistic threats in games with time order. Together these advances allow reliable predictions in practical fields such as auction design and legal institutions. The Nobel Committee honored them for establishing a framework that systematically explains rational outcomes even in non-cooperative environments.

Non-cooperative game theory studies strategic situations in which each agent seeks to maximize personal payoff while taking others’ actions as given. In the early 1950s Nash proved, using fixed-point theorems, that any finite game admits at least one equilibrium if mixed strategies are allowed. Yet the Nash equilibrium concept suffered from issues such as non-credible threats, time inconsistency, and lack of belief coherence under incomplete information. Selten refined the concept with “subgame perfect equilibrium,” requiring equilibrium conditions to hold in every subgame of the extensive form; the refinement has been applied to entry deterrence, fisheries agreements, and more. Harsanyi introduced player “types” as random variables assigned by “Nature” before play, creating a model of incomplete-information games and the Bayesian equilibrium, thereby unifying the analysis of negotiations, auctions, and public policy under asymmetric information. These theories laid the groundwork for mechanism design, forming the starting point for auction theory, contract theory, and experimental economics from the 1970s onward. Mathematical techniques such as fixed-point arguments, consistent Bayesian updating, and sequential rationality are now indispensable in building empirical models in industrial organization and political economy. The 1994 award is regarded as the decisive event that established game theory as a standard tool across economics.

Nash (1950, 1951) applied Brouwer and Kakutani fixed-point theorems to prove that the set of mixed strategy profiles in any finite n-player game is non-empty, compact, and convex, deriving equilibrium as a fixed point of the best-response correspondence. Selten (1965) analyzed extensive-form games via backward induction and imposed subgame consistency to eliminate non-credible threats, establishing the conceptual framework later known as Selten–Harsanyi perfection. His condition evolved into trembling-hand perfection, which selects equilibria robust to infinitesimal probabilities of player mistakes. Harsanyi’s 1967–1968 papers fully parameterized the type space, constructing hierarchical belief systems with a common prior and transforming games of incomplete information into games of complete but imperfect information. This enabled formalization of the inductive belief-updating process now called Bayesian proof. Their collective work stimulated the development of common-knowledge concepts and Aumann’s agreement theorem, forming the basis for equilibrium selection research. On the empirical side, experimental economics has tested trembling-hand and other refinement predictions, generating rich datasets that connect theory to observed behavior. In modern mechanism design, incentive compatibility and implementation problems are formalized starting from Nash equilibrium and its refinements. The 1994 award explicitly recognized these pioneering contributions that integrated refinement theory with information-structure analysis.