chapter  9
9 Pages

Probabilistic choice and strategic rationality

The idea of rational choice in uncertain conditions was elaborated in

Knight’s Risk, Uncertainty and Profit and Keynes’ Treatise on Probability, both published in 1921. Elements of risky decision making were present in

the works of Smith (1776: 120-122), Mill (1848: 407-409), Jevons (1871:

99) and Marshall (1890: 111).1 Yet, their theories were fundamentally

deterministic and none of them used probabilities as an instrument in eco-

nomic theorizing (Ménard 1978: 150; 1987: 141). John von Neumann and

Oskar Morgenstern (1944) introduced a rather new concept of rationality

associating utility maximization with probabilistic choice in risky situ-

ations. Leonard Savage moved forward this approach, explaining uncertain

prospects thanks to his “subjective expected utility” theory. John Nash

(1950) made the decisive step of suggesting an equilibrium solution to a

game with a finite number of players. Thus, interactive individuals parti-

cipate in non-cooperative games calculating the probabilities of possible

outcomes to maximize personal gains. With Nash, economics becomes a

logical science where agents with imperfect knowledge about the final

outcome make consistent yet interdependent choices. In this, two basic

limitations of the Walrasian/Paretian model are overcome, certainty and

automatic choice. In the latest versions of game theory, the idea of rational

strategic behaviour has introduced an extremely important learning process

in the cases of repeated and cooperative games. Rationality has turned out

to be not just a process of simply drawing logical conclusions about the

other player’s strategies, but also a process of continuous revision of the

interacting individual’s strategic beliefs.