This discussion paper resulted in a publication in the Journal of Economic Dynamics and Control 2013, 37(12), 2729-2754.
Existing models of R&D are not easily reconciled with four observable aspects of R&D: initial technologies ('ideas') need to be developed further, only a minority of initial ideas is successfully brought to the market, production and process innovations take place simultaneously (whereby, initially, there is no production at all), and process innovations are implemented for technologies that are destined to leave the market. We present a detailed bifurcation analysis for a dynamic model of R&D that captures these observations in one, unifying framework. As we provide a global analysis, we do not limit initial technologies to carry marginal costs that are below the choke price. We show that there always exists a critical value of initial marginal cost above which the firm does not initiate any (R&D) activity; the path to the saddle-point steady state is never globally optimal. We also sketch some tentative policy implications of our analysis.