Shill‐bidding in private values auctions

Abstract

Purpose

The purpose of this paper is to present implications of the seller's ability to bid in the four classical auction forms, with independent private values: English, Dutch, first‐ and second‐price auctions.

Design/methodology/approach

Under each auction form, the identity of the winning bidder and the expected winning bid are compared between the case when the seller may bid and when he cannot, using equilibrium bidder strategies. The seller's incentive to bid is evaluated.

Findings

The strategies and the welfare results differ with auction type and underlying information assumptions – bidders are either aware or unaware of the seller's ability to bid. In the Dutch and the first‐price auctions, seller‐bidding does not affect any classical results. In the English and the second‐price auctions, it leads to no lower expected prices than without it and higher prices with positive probability. In the English and the second‐price auctions, the seller bids above his reservation value and may unintentionally win the auction. These auctions result in inefficiency with positive probability.

Practical implications

The English and the second‐price auctions are the most common real‐world auctions. In these auctions, the seller's ability to bid – secretly or publicly – redistributes welfare among participants and introduces a possibility of inefficiency. Making this ability publicly known does not solve the latter problem. Auctioneers must prevent the seller from bidding, or must select a different auction form when seller‐bidding is anticipated.

Originality/value

The paper clarifies to regulators, auction designers, bidders, and other readers which auction forms are susceptible to subversion by seller‐bidding and what the potential damages are.

Citation

Hlasny, V. (2007), "Shill‐bidding in private values auctions", Journal of Information, Communication and Ethics in Society, Vol. 5 No. 4, pp. 307-320. https://doi.org/10.1108/14779960710846164