Undiscounted optimal growth in the two-sector Robinson-Solow-Srinivasan model: a synthesis of the value-loss approach and dynamic programming

We highlight two features of undiscounted optimal growth in the context of a two-sector model due to Robinson, Solow and Srinivasan. First, we use the value-loss approach of Radner-Gale-McKenzie to show a multiplicity of optimal programs in situations when optimality does not coincide with value-loss minimization. Second, we use a theory of undiscounted dynamic programming, not available in the literature, to derive properties of the optimal policy correspondence. In terms of a methodological perspective, we suggest a synthesis of the two methods for the analysis of problems of optimal intertemporal resource allocationThis essay is dedicated to Mukul Majumdar on the occasion of his sixtieth birthday, with affection and admiration. We would like to thank Robert Becker, Minako Fujio and Ron Jones for useful discussions and to a referee of this journal for very insightful comments. We are grateful to the Center for Analytic Economics at Cornell and to the Center for a Livable Future at Johns Hopkins for research support