In this paper, which forms a chapter in the forthcoming Book â Delivering a Low Carbon Electricity System: Technologies, Economics and Policyâ , Jamasb and Kohler revisit the literature on learning curves and their application to energy technology and climate change policy analysis and modeling. The academic literature and policy documents have in recent years embraced the learning curves and applied the concept to technology analysis and forecasting cost reductions. We argue that learning curves have often been used or assumed uncritically in technology analysis and draw parallels between the use of learning rates in energy technological progress and climate change modeling to that of discount rates in social cost benefit analysis. The paper discusses that care needs to be taken in applying learning curves, originally developed as an empirical tool to assess the effect of learning by doing in manufacturing, to analysis innovation and technical change. Finally, we suggest some potential extensions of learning curves, e.g. by incorporating R&D and diffusion effects into learning models, and other areas where learning curves may potentially be a useful tool in energy technology policy and analysis.