How to Set Minimum Acceptable Bids, with an Application to Real Estate Auctions

Abstract

Abstract:In a general auction model with affiliated signals, common components to valuations and endogenous entry, we compute the equilibrium bidding strategies and outcomes, and derive a lower bound on the optimal reserve price. This lower bound can be computed using data on past auction combined with information about the subsequent sales prices of unsold goods. We compute the lower bound using data from FDIC real estate auctions. We find the point estimate for an upper bound on the optimal reserve for real estate to be 75% of appraised value. Using data from private real estate auctions, we show that when reserve prices are around 50% of appraised value, seller revenues are increasing in the reserve price.

China's Land Market Auctions: Evidence of Corruption

“…7 We assume the reserve price is exogenous to the auction format choice, because it is set by an independent appraiser and empirically (below) it is not correlated with auction format. In a similar setting without corruption, McAfee, Quan, and Vincent (2002) analyze the setting of reserve price. With endogenous reserve price, our corruption story is likely strengthened because the corrupt official is more likely to lower reserve price for two-stage auction, and the expected revenue loss to the city would be larger.…”

Econometrics of first-price auctions with entry and binding reservation prices

“…Also, whether the bids are optimal according to the Nash-Bayesian equilibrium strategies can be tested. An interesting extension would be to drop the assumption that the active bidders learn the number of active bidders n prior to stage 2 and to allow the active bidders to bid without knowing n: See, e.g., the theoretical model developed in McAfee et al (2002) to study optimal reserve prices in real estate auctions and the nonparametric analysis of a similar model within a common value paradigm using the outer continental shelf (OCS) auctions data by Hendricks et al (2003). While the theoretical analysis from a game-theoretic viewpoint yields the same results whether or not the active bidders know n before they bid, the econometric implementation is more involved.…”

Efficient investment in a dynamic auction environment

“…This necessitates public messages within the mechanism, rather than messages sent only to the principal/designer/seller. 2 Entry costs are investigated in various contexts by Milgrom (1981Milgrom ( , 2004, Samuelson (1985), McAfee and McMillan (1987), Riordan and Sappington (1987), Levin and Smith (1994), Fullerton and McAfee (1999), McAfee et al (2002), and Ye (2004Ye ( , 2007.…”