Does it Pay Off to Bid Aggressively? An Empirical Study Philipp Herrmann University of Paderborn philipp.herrmann@wiwi.upb.de Dennis Kundisch University of Paderborn dennis.kundisch@wiwi.upb.de Mohammad S. Rahman University of Calgary rahman@ucalgary.ca Abstract We empirically investigate the payoff of signaling through aggressiveness in an online auction. To address our research question, we use a unique and very rich dataset containing actual market transaction data for approximately 7,000 pay-per-bid auctions. Our research design allows us to isolate the impact of aggressive bidding, used in an attempt to signal a high valuation to deter other auction participants, on the probability of winning an auction. We analyze more than 600,000 bids placed manually by approximately 2,600 distinct auction participants. We find a strong and significant positive effect of aggressive bidding on the total number of bids placed, and on the total number of participants in an auction. The strong and significantly negative effect of aggressive bidding on the individual probability of winning an auction supports the finding that aggressive bidding is ineffective as a strategy for deterring competitors in an online auction. 1. Introduction “… he bid seventy-five grand for the land when the other operators were offering bids in the low fifties…. Naturally he got it … and made himself a sweet little bundle. After he bought it, I told him he could have got it for twenty thousand less and you know what he said? ‘I never try to buy a property as cheap as possible. That way you’re in competition with the other operators. They keep kicking each other up and before you know it, you’re paying more than you intended and more than it’s worth to me, and that’s what I offer. That way you discourage the competition. It takes the heart right out of him.” –Harry Kemelman, Wednesday the Rabbi Got Wet. [3] Online auctions have become a mainstream economic phenomenon. For example, in 2011, the total value of goods sold on eBay – one of the biggest online auction websites worldwide – was $68.6 billion. It is hardly surprisingly, then, that for over a decade online auctions have featured prominently as a study topic in the IS literature (e.g., [7], [17]) as well as in the economics literature (e.g., [29], [25]). One central topic in this literature concerns the analysis of different types of bidders and bidding strategies – such as aggressive bidding – in these auctions (e.g., [9], [6], [23]). The quote by Harry Kemelman may be taken as anecdotal evidence that an aggressive bidding strategy promises a positive return for the respective bidder. However, Kemelman cannot be sure that he won because of his aggressive bidding strategy. It is also possible that he is simply the bidder with the highest valuation of the auctioned product and, thus, could also have won with a lower bid using a more cautious bidding strategy. The existence of aggressive bidding strategies – typically called jump bidding – in ascending price (online) auctions has been documented extensively in the literature (e.g., [3], [14], [12], [20]). Cramton [11] defines jump bidding as “the act of raising a high bid by much more than the minimum increment”. Empirical studies suggest that jump bidding occurs quite frequently. For example, Easley and Tenorio [14] report that more than 30% of bidders in their sample submitted jump bids. Naturally, a significant literature has emerged, analyzing these strategies theoretically and empirically (for an extensive literature review see, e.g., [28]). In general, signaling (e.g., [3], [12]) and impatience (e.g., [20]) have been named as potential explanations for jump bidding. However, none of the existing studies empirically investigated the impact of jump bids as a way of signaling a high valuation on the actual likelihood of winning an auction. Thus, the question, whether Kemelman won the auction because of the effectiveness of his bidding strategy or merely because of his high valuation of the land remains until today unanswered. This gap in the literature may be explained by two important challenges: First, the distinction between aggressive bids that are attributable to impatience and aggressive bids that are attributable to signaling is hard to make. Second, and more importantly, even when impatience is ruled out as a cause for aggressive bids, in typical ascending price auctions these bids may imply two simultaneous effects: (1) By definition, an aggressive bid always increases the auction price by 2013 46th Hawaii International Conference on System Sciences 1530-1605/12 $26.00 © 2012 IEEE DOI 10.1109/HICSS.2013.183 3960 2013 46th Hawaii International Conference on System Sciences 1530-1605/12 $26.00 © 2012 IEEE DOI 10.1109/HICSS.2013.183 3962