Optimal Mechanism Design for Procurement Auctions for Public Works in Mumbai

We examine procurement auctions for public works in Mumbai with the goal of understanding game theoretic values and optimal auction structures to recommend to policy makers in this context. We first convert an independent private values framework for traditional auctions to an analogous framework for procurement auctions. Then we use a fixed effects regression to estimate the idiosyncratic components of bids, which we use to estimate bidders’ private value distribution. Finally we then estimate the optimal auction reservation values. Since today these auctions do not include reservation values, we conclude that using these optimal reservation values would likely reduce expected costs for public works tenders in Mumbai.