Efforts to estimate revenues from future spectrum auctions are fraught with difficulties. Revenues depend heavily on specific auction characteristics, such as bidding structures and the geographic features of the licenses at auction, making it difficult to use past results to predict future outcomes. Predicting revenues from incentive auctions raises even greater challenges, since the net revenues received by the government depend on the nature and magnitude of the “incentives” offered to current licensees. Based on the information currently available about the FCC’s proposal to use incentive auctions to repurpose spectrum for mobile wireless use, the revenues that might be produced by such an auction are unknowable with any degree of precision.