In this paper, we investigate the problem of the strategic foundation of the
Cournot–Walras equilibrium approach. To this end, we respecify à la Cournot–Walras
the mixed version of a model of simultaneous, noncooperative exchange, originally
proposed by Lloyd S. Shapley. We show, through an example, that the set of the
Cournot–Walras equilibrium allocations of this respecification does not coincide with
the set of theCournot–Nash equilibrium allocations of the mixed version of the original
Shapley’s model. As the nonequivalence, in a one-stage setting, can be explained by
the intrinsic two-stage nature of the Cournot–Walras equilibrium concept, we are led
to consider a further reformulation of the Shapley’s model as a two-stage game, where
the atoms move in the first stage and the atomless sector moves in the second stage.
Our main result shows that the set of the Cournot–Walras equilibrium allocations coincides
with a specific set of subgame perfect equilibrium allocations of this two-stage
game, which we call the set of the Pseudo–Markov perfect equilibrium allocations.
