Researchers propose a solution to 'shill-bidding'
Researchers at the University of Southampton have proposed a solution to 'shill-bidding', a disruptive but increasingly prevalent practice which threatens the integrity of online auctions such as eBay.
Shill-bidding occurs when a seller, or a friend or associate of a seller, bids clandestinely on their own sale. Sellers can thus increase the price of their own items or ultimately buy them back if the auction fails to achieve the price they want.
Recent media articles have highlighted the practice and the problems it poses for online auction sites (Sunday Times, 28/1/07 http://www.timesonline.co.uk/article/0,,2087-2570548.html. Not only does it artificially inflate the sale price, causing financial loss to potential buyers, but it also erodes faith in the integrity and fairness of the marketplace. The practice is illegal in many countries and prohibited on most common online auction sites such as eBay and Amazon.
However, shill-bidding is notoriously hard to detect and harder still to substantiate, particularly in online auctions where it is easy for bidders to use false identities.
Now researchers in the School of Electronics and Computer Science who are at the forefront of auction mechanism research (Dr Enrico Gerding, Dr Alex Rogers, Dr Rajdeep Dash and Professor Nicholas Jennings) have proposed an effective and simple way of combating shilling, which avoids legislation. In a paper presented to the world's leading conference on artificial intelligence in India this month, they suggest using different kinds of listing and commission fees from those currently in use on eBay.
Shill-bidding is particularly relevant to eBay since, in addition to a percentage of the selling price, eBay also charges a fee if the seller sets a minimum price in the form of a reserve or starting price. However, sellers can avoid this fee by placing a shill bid instead, thereby effectively 'hiding' their minimum price. Moreover, according to eBay, 'sellers have found that setting the starting price too high may discourage bidding.' As a result, shill-bidding becomes attractive to the seller since it provides a way of setting reserve/starting prices surreptitiously.
The alternative fee structure proposed by the ECS researchers takes a percentage only of the difference between the set minimum price and the final selling price. This fee effectively rewards the setting of a minimum price and has been shown to reduce the incentive for sellers to shill bid, as well as providing higher benefits to the buyers.
Notes for editors
The paper was presented at the International Joint Conference on Artificial Intelligence (IJCAI 2007) in Hyderabad in January.