OPTIMAL-GROWTH AND PARETO OPTIMALITY

The purpose of this paper is to show that in a stationary intertemporal economy where agents have recursive utilities every Pareto optimum is a solution of a generalized McKenzie problem. An 'abstract' state space is introduced as the space of couples of capital stock and utilities that can be reached by n-1 agents from that capital stock. 'Generalized technological' conditions are then defined on that abstract space as well as a recursive criterion on sequences of its elements. The criterion generalizes the additively separable one. As Bellman's and Euler's equations still hold, many dynamical results known for the additively separable one-agent case can be generalized.