Who is more committed than me? A dynamic structural model of bidder strategies in online auctions

In online auctions, bidders frequently use an aggressive bidding strategy. Although numerous studies have examined its motivations and economic consequences, relatively few have explored aggressive bidding as a dynamic decision process and how reputation affects future bidding. Two challenges make it difficult to empirically estimate this effect using archival data. First, there exist confounding effects such as sunk cost and inertia. Second, aggressive bidding is an endogenous process. To fill this gap, we develop a model to separately identify the effect of aggressive bidding from the other two confounding factors. Our reduced-form model results show that we successfully differentiate the effects by all three factors. More importantly, the positive effect of reputation confirms that aggressive bidding promotes subsequent bidding. Our results imply that how the auction website provides information feedback has a critical impact on bidding dynamics. This study will make contributions beyond the immediate research context.