In this talk I’ll discuss two projects investigating policy to promote wind energy. The first paper (draft attached) asks whether it is better to subsidize the construction of wind farms (investment) or to subsidize each unit of power wind farms generate (output). Leveraging a natural experiment which unexpectedly gave wind developers a choice between these two subsidy options, we find that shifting subsidies from the output to the investment margin caused a ten percent reduction in wind farm productivity.

In the second part of the talk, I’ll discuss early stage work investigating the implications of technological improvement for policy design. We first derive a simple “engineering”-based way of characterizing which projects should wait for better technology. We then introduce endogenous technological change (generated by manufacturer experience), and discuss ways in which the observed ordering of installations may deviate from the social optimum. Finally, we present preliminary evidence on the extent of wind turbine misallocation from a database of over 8,000 wind farms worldwide.