Best practice of learning curves.

Learning curves are no doubt very useful for understanding matters of technology development and its consequences. But could they also be used for policy design? The answer is yes and in many ways. It has further been done so in some policy design such as the Japanese million-roof programme for Photovoltaics and the German feed-in tariffs for renewable fuels. It can be questioned if the policy makers have acted strictly on analytical basis or rather on intuition, but there is evidence that they in the process have been made aware of what they can expect, in terms of lowering of costs and in improved performance, as the cumulative stock of technology on the market grows.

It may also for the policy maker be useful to know that some subsidies (that their opponents are arguing to be uselessly spent) could be looked upon as “learning investments”. Because that is the case when a technology supported financially “rides down the learning curve” and yields profit when it overtakes the incumbent technology as best economic choice. It would be further comforting for the policy maker to know that if (and when) the nichemarkets can be identified the learning investments can be shared and that government can get away with less for starting the useful learning process. See pictures below and also this paper.