THE DETERMINANTS OF SNIPING ON eBAY: AN ECONOMETRIC ANALYSIS

eBay is the largest and most widely known of many internet auction sites designed to connect buyers and sellers around the world. An eBay auction consists of a set bidding period during which individuals submit “maximum bids” for an item. The winner the auction is the individual with the highest bid, and the price is equal to the second highest bid, plus a small bidding increment. This system is a close approximation to a second-price auction, for which William Vickrey proved in 1961 that a weakly dominant strategy is to bid one’s own private value for an item. A bidding strategy called “sniping,” whereby an individual withholds a bid until the last moment of the bidding period, is widely observed on eBay, but seems to violate rational behavior in the Vickrey model. Bidding late should not have any advantage over bidding early in a second-price auction, so economists have sought ways to rationalize its existence. In this paper, I create a data set from 515 eBay auctions and test several potential determinants of sniping behavior using a standard linear regression model. I incorporate a number of variables from previous literature (a common value indicator for an item, the feedback rating of an individual, and the number of opponents in an auction), as well as a new variable (the number of substitutes for an item), and achieve significant results.