Abstract

The paper contains first the analysis of the strategic decision as to whether to integrate or not, in a market with differenciation of final products feeding back into the production of intermediate inputs. Cournot competition makes integration a dominant strategy, althogh it is not Pareto optimal for the industry when goods are close substitutes.Bertrand competition leaves room also for non integration and there remains the possibility of asymmetric industry organization, with non integrated firms competing with integrated rivals.The analysisis extended to an oligopoly where upstream process R&D takes place. Here, the nonintegrated part of industry may invest more in process R&D and even perform better than its integrated counterart.