Research

Abstract: Auctions for large-scale procurement of electricity production capacity have become important policy tools for achieving energy security and improving electricity access in developing countries. These auctions have procurement targets that often exceed any supplier's capacity, resulting in nuanced allocation rules. In this paper, I study allocation efficiency and auctioneer's payoff in these auctions and suggest ways to improve them. I use India as my leading example, where a non-trivial variant of US treasury auctions has helped create 140 gigawatts of capacity, which is fourth largest globally. I theoretically prove that the design of these auctions incentivises lower-capacity firms to be more competitive. Such firms may get larger contracts despite having higher costs, leading to inefficient allocation. To measure the inefficiency, I first estimate a structural model to find firms' cost distributions using a recent dataset. Then, I simulate the theoretical equilibrium with estimated cost distributions to quantify the inefficiency. I further suggest counterfactual designs which lower the inefficiency and government expenditure, without reducing capacity allocation.