Do Consumers Discount Parallel Imports?

Using data from Taiwan's Yahoo! auctions of Nikon cameras, this paper investigates whether there exists any difference in transaction results between commodities which are sold by authorized sellers and those which are parallel imports. We find that the parallel imports and the authorized products differ not only in the warranty duration offered by the sellers, but also in the formats in which they are listed, both of which substantially affect trade probability and transaction price. We first use the sellers' inventory and experience as instrumental variables to endogeneize their choice of the fixed-price listing and BIN auction, respectively. We show that, after taking into account the endogenous choice of listing formats and the characteristics of the sellers and the items, an authorized product has a 7 percent higher trade probability than that of a parallel import and, if there is a sale, the transaction price as a ratio of the manufacturer's suggested retail price is 0.093 higher. A small number of less experienced parallel import sellers, in an attempt to maintain a price comparable to the authorized product, list their items in pure auction with unusually high starting bids. This accounts for the overall lower trade probability for the parallel imports.If we disregard these sellers who adopt pure auction, the authorized product has higher transaction price than, but about the same trade probability as the parallel import