Comparing the Spanish and the discriminatory auction formats: A discrete model with private information

Departamento de Economía Cuantitativa, Universidad Complutense, Campus de Somosaguas, 28223 Madrid, Spain
European Journal of Operational Research (Impact Factor: 2.36). 02/2007; 179(1):253-266. DOI: 10.1016/j.ejor.2006.02.044


The Spanish Treasury is the only Treasury in the world that uses a hybrid system of discriminatory and uniform price auctions to sell government debt: winning bidders pay their bid price for each unit if this is lower than the weighted average price of winning bids (WAP), and pay the WAP otherwise. Following Gordy [Gordy, M., 1996. Multiple bids in a multiple-unit common-value auction. Board of Governors of the Federal Reserve System], we model the Spanish auction as a common value auction of multiple units with private information, allowing for multiple bids. Numerical analysis shows that bidders spread their bids more in the Spanish than in the discriminatory auction and bid higher for the first unit, and that the expected seller’s revenue is higher in the Spanish than in the discriminatory auction within a reasonable set of parameter values.

24 Reads
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We study dominant strategy implementation in a variant of the canonical public good provision model, as proposed by Borgers and Postl (2009). In this set up, we fully characterize the set of budget-balanced dominant strategy deterministric mechanisms, which are simple threshold rules. For probabilistic mechanisms that are continuously differentiable we provide a necessary and sufficient condition for dominant strategy implementation. When allowing for discontinuities in the mechanism, our necessary condition remains valid, but additional requirements must be met in order to ensure sufficiency.
    Portuguese Economic Journal 04/2009; 8(1):3-14. DOI:10.1007/s10258-009-0042-z · 0.12 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper investigates how economic outcomes differ for new model introduction between two extreme contracts: outright sales contract and returns policy contract with full credit. Under the outright sales contract, we show that the retailer's incentive to introduce a new model can lead to a reduction of consumer welfare when the retailer is moderately risk averse. The primary conclusion from this study is that the returns policy contract resolves this conflict between firms and consumers, and that the supply chain introduces the new model only when consumer surplus from the new model is higher than that from the old model.
    International Journal of Production Economics 04/2010; 124(2-124):299-309. DOI:10.1016/j.ijpe.2009.10.009 · 2.75 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper examines the practice of phantom bidding in common value auctions with asymmetric information. Unlike in private value auctions it is very difficult to profit from phantom bidding even if the auctioneer observes the number of interested bidders. We find that the auctioneer is worse off submitting low bids or mixing between his participation strategies. Phantom bidding can generate higher revenues if a single bidder with a high value estimate is likely to be present at the auction. Systematic observation of an isolated high estimate requires negative correlation across bidders’ information that is atypical in a common value environment.
    International Journal of Economic Theory 05/2011; 7(2):217 - 230. DOI:10.1111/j.1742-7363.2011.00160.x · 0.38 Impact Factor