Walras' theory of general economic equilibrium undergoes deep changes over the twenty-six years elapsing from the date when Walras' first significant theoretical contribution is published (1874) and the date when the fourth edition of his fundamental work, the Elements d'economie politique pure, eventually appears (1900). In this paper we reconstruct the many steps characterizing the twisted evolution of Walras' thought concerning equilibrium, disequilibrium, and time in economics. Our reconstruction shows how difficult it is for Walras to free his analysis from the constraints initially imposed on it by a pre-analytic vision marked by a strong, almost naif, descriptive realism, at least as far as the representation of the competitive market mechanism is concerned. As a matter of fact, Walras very early, albeit only confusedly, arrives at perceiving the theoretical, analytical and epistemological difficulties undermining his analysis of the tâtonnement processes associated to his various equilibrium models, but he hesitates to face such issues squarely lest this would force him to give up the allegedly realistic justifications of his theory. As far as the pure exchange model is concerned, Walras, under the pressure of external criticism, soon enough (i.e., since 1885) yields to the requirements of the analysis, by explicitly endorsing the so-called no-trade-out-of-equilibrium assumption, as well as the associated interpretation of the tâtonnement process in exchange as a purely virtual equilibration process in time, supporting an instantaneous equilibrium notion. As far as the production models are concerned, on the contrary, Walras strenuously resists giving up the pseudo-realistic assumptions permeating his analysis of the tâtonnement process in production at least up to the third edition of the Elements (1896): such analysis is essentially based on the idea that the tâtonnement process in production should be interpreted as actual equilibration process in real time, supporting a stationary equilibrium notion. Yet, due to his late realization of the logical contradictions engendered by his assumptions concerning this sort of tâtonnement process and equilibrium notion, hardly (if at all) compensated by a few dubious gains in terms of descriptive realism, Walras eventually resolves, in the fourth edition of the Elements, to renounce all pseudo-realistic pretence: in the 1900 edition, in fact, he adopts a novel, openly unrealistic assumption, the so-called hypothese des bons, which turns all tâtonnement processes, including the one concerning production, into virtual processes in time, supporting a generalized notion of instantaneous equilibrium of the temporary type.
Read full abstract