资源环境科技发展态势分析平台

Uncertainty in technological learning is a crucial factor in planning research, development, and demonstration (RD&D) strategies. Nevertheless, most previous work either models technological change as deterministic or accounts for uncertainty without fully capturing the recourse feature of the problem. This paper improves upon these approaches by developing a real options-based stochastic dynamic programming method for valuing and planning low-carbon energy RD&D investment and is the first of its kind to disaggregate the effects of R&D and learning-by-doing. This simplified model captures the relevant features of the problem and provides general insights on RD&D strategy under technological uncertainty. Results indicate that imminent deployment, high cost, lower exogenous cost reductions, and lower program funds all promote R&D spending over learning-by doing, since under these circumstances a breakthrough, rather than slow and consistent cost reductions, will render the program successful.