Optimal Diversity in Auctions for Renewable Energy Sources under Technological Uncertainty

A cost-effective low-carbon transition requires designing a state support mechanism that delivers an optimal diversity of renewable energy sources (RES) in the energy mix. Lowest price auctions that do not discriminate between technologies deliver optimal RES diversity, providing that there are no spill-over effects associated with the deployment of each technology. This precondition fails to apply, however, if RES technologies are able to benefit from learning-by-doing and if learning rates are uncertain. In the first part of this study we review the literature on the optimal diversity of technologies when technological progress is uncertain and on the uncertainty of learning rates. Then we use an analytical model to demonstrate that, under the uncertainty of learning potential, the socially-optimal diversity of RES is larger than the outcome of the lowest price auction. We also show that the social benefits from diversification disappear if there is no potential for learning-by-doing. Thus, countries that potentially could benefit from large learning rate effects — such as countries at the technological frontier — should increase RES diversification by introducing technology-specific auctions, while more peripheral countries should limit diversification by using technology-neutral auctions. We also show that the diversity of RES in the social optimum is greater than that predicted by energy models assuming fixed learning rates.