Cet article constitue la mise au point et le developpement d'une communication anterieure sur le meme sujet. Il concerne le probleme pose par l'incompatibilite de la regle de vente au cout marginal preconisee par les Welfare Economics, et les conditions d'equilibre budgetaire imposees par les Pouvoirs Publics aux entreprises nationalisees. La methode consiste en une de Pareto appliquee a un modele general dont les liaisons de structure comportent a la fois des liaisons entre les quantites et des liaisons entre les valeurs. Les resultats obtenus, rassembles dans la section 8, sont commentes et compares aux solutions anterieurement proposees a ce probleme. Un resume de cet article est constitue par les sections 1, 3, et 8, qui peuvent etre lues sans prendre connaissance des calculs. Le modele developpe dans la section 2 a pour objet de faciliter l'interpretationdonnee dans la section 3, et d'expliquer plus clairement l'intervention des prix et revenus dans les liasons du systeme; mais la maximation exposee dans la section 5 est effectuee sur le modele transforme qui, a la fin de la section 4, rassemble sous une forme condensee les liaisions du systeme. La section 9, enfin, a l'occasion de la critique faite de la politique traditionnelle des monopoles publics, commente certains points particuliers des resultats exposes dans la section 8.

following study proposes to generalize the classical theory of economic equilibrium and of welfare economics to the case of risk. Besides our own work, the only similar attempt which has been made in this field is, to the best of our knowledge, Mr. Arrow's paper, The Role of Securities in the Optimal Allocation of Risk-Bearing, presented at the international colloquium on risk which was held in Paris, May, 1952. It is quite astonishing that such an important chapter of economics, both from theoretical and practical points of view, has up to now never been the object of systematic study. reason for this is certainly its great difficulty. As the method of attacking this problem is doubtless the basic difficulty, we have, in order to make the exposition easier, confined ourselves to a simplified model. This has the double advantage of making perfectly evident the main considerations which, in our opinion, must enter into the analysis and of providing a simple and explicit framework for discussion. This model, furthermore, lends itself to easy generalization. Generally it can be said that the essential results of the theories of economic equilibrium and welfare economics are still valid in the case of risk.